To see the other types of publications on this topic, follow the link: Black-box learning algorithm.
Journal articles on the topic 'Black-box learning algorithm'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Black-box learning algorithm.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Hwangbo, Jemin, Christian Gehring, Hannes Sommer, Roland Siegwart, and Jonas Buchli. "Policy Learning with an Efficient Black-Box Optimization Algorithm." International Journal of Humanoid Robotics 12, no. 03 (September 2015): 1550029. http://dx.doi.org/10.1142/s0219843615500292.
Full textAbstract:
Robotic learning on real hardware requires an efficient algorithm which minimizes the number of trials needed to learn an optimal policy. Prolonged use of hardware causes wear and tear on the system and demands more attention from an operator. To this end, we present a novel black-box optimization algorithm, Reward Optimization with Compact Kernels and fast natural gradient regression (ROCK⋆). Our algorithm immediately updates knowledge after a single trial and is able to extrapolate in a controlled manner. These features make fast and safe learning on real hardware possible. The performance of our method is evaluated with standard benchmark functions that are commonly used to test optimization algorithms. We also present three different robotic optimization examples using ROCK⋆. The first robotic example is on a simulated robot arm, the second is on a real articulated legged system, and the third is on a simulated quadruped robot with 12 actuated joints. ROCK⋆ outperforms the current state-of-the-art algorithms in all tasks sometimes even by an order of magnitude.
2
Kirsch, Louis, Sebastian Flennerhag, Hado van Hasselt, Abram Friesen, Junhyuk Oh, and Yutian Chen. "Introducing Symmetries to Black Box Meta Reinforcement Learning." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 7 (June 28, 2022): 7202–10. http://dx.doi.org/10.1609/aaai.v36i7.20681.
Full textAbstract:
Meta reinforcement learning (RL) attempts to discover new RL algorithms automatically from environment interaction. In so-called black-box approaches, the policy and the learning algorithm are jointly represented by a single neural network. These methods are very flexible, but they tend to underperform compared to human-engineered RL algorithms in terms of generalisation to new, unseen environments. In this paper, we explore the role of symmetries in meta-generalisation. We show that a recent successful meta RL approach that meta-learns an objective for backpropagation-based learning exhibits certain symmetries (specifically the reuse of the learning rule, and invariance to input and output permutations) that are not present in typical black-box meta RL systems. We hypothesise that these symmetries can play an important role in meta-generalisation. Building off recent work in black-box supervised meta learning, we develop a black-box meta RL system that exhibits these same symmetries. We show through careful experimentation that incorporating these symmetries can lead to algorithms with a greater ability to generalise to unseen action & observation spaces, tasks, and environments.
3
Xiang, Fengtao, Jiahui Xu, Wanpeng Zhang, and Weidong Wang. "A Distributed Biased Boundary Attack Method in Black-Box Attack." Applied Sciences 11, no. 21 (November 8, 2021): 10479. http://dx.doi.org/10.3390/app112110479.
Full textAbstract:
The adversarial samples threaten the effectiveness of machine learning (ML) models and algorithms in many applications. In particular, black-box attack methods are quite close to actual scenarios. Research on black-box attack methods and the generation of adversarial samples is helpful to discover the defects of machine learning models. It can strengthen the robustness of machine learning algorithms models. Such methods require queries frequently, which are less efficient. This paper has made improvements in the initial generation and the search for the most effective adversarial examples. Besides, it is found that some indicators can be used to detect attacks, which is a new foundation compared with our previous studies. Firstly, the paper proposed an algorithm to generate initial adversarial samples with a smaller L2 norm; secondly, a combination between particle swarm optimization (PSO) and biased boundary adversarial attack (BBA) is proposed. It is the PSO-BBA. Experiments are conducted on the ImageNet. The PSO-BBA is compared with the baseline method. Experimental comparison results certificate that: (1) A distributed framework for adversarial attack methods is proposed; (2) The proposed initial point selection method can reduces query numbers effectively; (3) Compared to the original BBA, the proposed PSO-BBA algorithm accelerates the convergence speed and improves the accuracy of attack accuracy; (4) The improved PSO-BBA algorithm has preferable performance on targeted and non-targeted attacks; (5) The mean structural similarity (MSSIM) can be used as the indicators of adversarial attack.
4
LIU, Yanhe, Michael AFNAN, Vincent CONTIZER, Cynthia RUDIN, Abhishek MISHRA, Julian SAVULESCU, and Masoud AFNAN. "Embryo Selection by “Black-Box” Artificial Intelligence: The Ethical and Epistemic Considerations." Fertility & Reproduction 04, no. 03n04 (September 2022): 147. http://dx.doi.org/10.1142/s2661318222740590.
Full textAbstract:
Background: The combination of time-lapse imaging and artificial intelligence (AI) offers novel potential for embryo assessment by allowing a vast quantity of image data to be analysed via machine learning. Most algorithms developed to date have used neural networks which are uninterpretable (“black-box”) and cannot be understood by doctors, embryologists and patients, which raises ethical and epistemic concerns for embryo selection in a clinical setting. Aim: This study aims to discuss ethical and epistemic considerations surrounding clinical implementation of “black-box” based embryo selection algorithms. Method: A scoping review was performed by evaluating publications reporting “black-box” embryo selection algorithms. Potential ethical and epistemic issues were identified and discussed. Results: No randomised controlled trial was identified in the literature evaluating clinical effectiveness of “black-box” embryo selection algorithms. Several ethical and epistemic concerns were identified. Potential ethical issues included (1) lack of randomised controlled trials, (2) impact on the shared decision-making process in embryo selection between clinicians and patients, (3) misrepresentation of patient values due to hidden reasoning process in “black-box” algorithms, (4) social impacts if algorithm subsequently proven to be biased, and (5) unclear responsibility when algorithm makes obviously poor choices of embryos. Potential epistemic issues included (1) information asymmetries between algorithm developers and doctors, embryologists and patients; (2) risk of biased prediction due to data selection during training process; (3) inability to troubleshoot for data training purposes due to limited interpretability; and (4) the economics of buying into commercial proprietary add-ons. Conclusion: There are significant epistemic and ethical concerns with “black-box” embryo selection. No published randomised controlled trial is available to support its clinical implementation. AI embryo selection in general, however, is potentially useful but must be done carefully and transparently. Interpretable AI would be preferred alternative in causing fewer issues.
5
Bausch, Johannes. "Fast Black-Box Quantum State Preparation." Quantum 6 (August 4, 2022): 773. http://dx.doi.org/10.22331/q-2022-08-04-773.
Full textAbstract:
Quantum state preparation is an important ingredient for other higher-level quantum algorithms, such as Hamiltonian simulation, or for loading distributions into a quantum device to be used e.g. in the context of optimization tasks such as machine learning. Starting with a generic "black box" method devised by Grover in 2000, which employs amplitude amplification to load coefficients calculated by an oracle, there has been a long series of results and improvements with various additional conditions on the amplitudes to be loaded, culminating in Sanders et al.'s work which avoids almost all arithmetic during the preparation stage.In this work, we construct an optimized black box state loading scheme with which various important sets of coefficients can be loaded significantly faster than in O(N) rounds of amplitude amplification, up to only O(1) many. We achieve this with two variants of our algorithm. The first employs a modification of the oracle from Sanders et al., which requires fewer ancillas (log2⁡g vs g+2 in the bit precision g), and fewer non-Clifford operations per amplitude amplification round within the context of our algorithm. The second utilizes the same oracle, but at slightly increased cost in terms of ancillas (g+log2⁡g) and non-Clifford operations per amplification round. As the number of amplitude amplification rounds enters as multiplicative factor, our black box state loading scheme yields an up to exponential speedup as compared to prior methods. This speedup translates beyond the black box case.
6
MIKE, KOBY, and ORIT HAZZAN. "MACHINE LEARNING FOR NON-MAJORS: A WHITE BOX APPROACH." STATISTICS EDUCATION RESEARCH JOURNAL 21, no. 2 (July 4, 2022): 10. http://dx.doi.org/10.52041/serj.v21i2.45.
Full textAbstract:
Data science is a new field of research, with growing interest in recent years, that focuses on extracting knowledge and value from data. New data science education programs, which are being launched at a growing rate, are designed for multiple levels, beginning with elementary school pupils. Machine learning is an important element of data science that requires an extensive background in mathematics. While it is possible to teach the principles of machine learning as a black box, it might be difficult to improve algorithm performance without a white box understanding of the underlaying learning algorithms. In this paper, we suggest pedagogical methods to support white box understanding of machine learning algorithms for learners who lack the needed graduate level of mathematics, particularly high school computer science pupils.
7
García, Javier, Roberto Iglesias, Miguel A. Rodríguez, and Carlos V. Regueiro. "Directed Exploration in Black-Box Optimization for Multi-Objective Reinforcement Learning." International Journal of Information Technology & Decision Making 18, no. 03 (May 2019): 1045–82. http://dx.doi.org/10.1142/s0219622019500093.
Full textAbstract:
Usually, real-world problems involve the optimization of multiple, possibly conflicting, objectives. These problems may be addressed by Multi-objective Reinforcement learning (MORL) techniques. MORL is a generalization of standard Reinforcement Learning (RL) where the single reward signal is extended to multiple signals, in particular, one for each objective. MORL is the process of learning policies that optimize multiple objectives simultaneously. In these problems, the use of directional/gradient information can be useful to guide the exploration to better and better behaviors. However, traditional policy-gradient approaches have two main drawbacks: they require the use of a batch of episodes to properly estimate the gradient information (reducing in this way the learning speed), and they use stochastic policies which could have a disastrous impact on the safety of the learning system. In this paper, we present a novel population-based MORL algorithm for problems in which the underlying objectives are reasonably smooth. It presents two main characteristics: fast computation of the gradient information for each objective through the use of neighboring solutions, and the use of this information to carry out a geometric partition of the search space and thus direct the exploration to promising areas. Finally, the algorithm is evaluated and compared to policy gradient MORL algorithms on different multi-objective problems: the water reservoir and the biped walking problem (the latter both on simulation and on a real robot).
8
Mayr, Franz, Sergio Yovine, and Ramiro Visca. "Property Checking with Interpretable Error Characterization for Recurrent Neural Networks." Machine Learning and Knowledge Extraction 3, no. 1 (February 12, 2021): 205–27. http://dx.doi.org/10.3390/make3010010.
Full textAbstract:
This paper presents a novel on-the-fly, black-box, property-checking through learning approach as a means for verifying requirements of recurrent neural networks (RNN) in the context of sequence classification. Our technique steps on a tool for learning probably approximately correct (PAC) deterministic finite automata (DFA). The sequence classifier inside the black-box consists of a Boolean combination of several components, including the RNN under analysis together with requirements to be checked, possibly modeled as RNN themselves. On one hand, if the output of the algorithm is an empty DFA, there is a proven upper bound (as a function of the algorithm parameters) on the probability of the language of the black-box to be nonempty. This implies the property probably holds on the RNN with probabilistic guarantees. On the other, if the DFA is nonempty, it is certain that the language of the black-box is nonempty. This entails the RNN does not satisfy the requirement for sure. In this case, the output automaton serves as an explicit and interpretable characterization of the error. Our approach does not rely on a specific property specification formalism and is capable of handling nonregular languages as well. Besides, it neither explicitly builds individual representations of any of the components of the black-box nor resorts to any external decision procedure for verification. This paper also improves previous theoretical results regarding the probabilistic guarantees of the underlying learning algorithm.
9
Anđelić, Nikola, Ivan Lorencin, Matko Glučina, and Zlatan Car. "Mean Phase Voltages and Duty Cycles Estimation of a Three-Phase Inverter in a Drive System Using Machine Learning Algorithms." Electronics 11, no. 16 (August 21, 2022): 2623. http://dx.doi.org/10.3390/electronics11162623.
Full textAbstract:
To achieve an accurate, efficient, and high dynamic control performance of electric motor drives, precise phase voltage information is required. However, measuring the phase voltages of electrical motor drives online is expensive and potentially contains measurement errors, so they are estimated by inverter models. In this paper, the idea is to investigate if various machine learning (ML) algorithms could be used to estimate the mean phase voltages and duty cycles of the black-box inverter model and black-box inverter compensation scheme with high accuracy using a publicly available dataset. Initially, nine ML algorithms were trained and tested using default parameters. Then, the randomized hyper-parameter search was developed and implemented alongside a 5-fold cross-validation procedure on each ML algorithm to find the hyper-parameters that will achieve high estimation accuracy on both the training and testing part of a dataset. Based on obtained estimation accuracies, the eight ML algorithms from all nine were chosen and used to build the stacking ensemble. The best mean estimation accuracy values achieved with stacking ensemble in the black-box inverter model are R¯2=0.9998, MAE¯=1.03, and RMSE¯=1.54, and in the case of the black-box inverter compensation scheme R¯2=0.9991, MAE¯=0.0042, and RMSE¯=0.0063, respectively.
10
Veugen, Thijs, Bart Kamphorst, and Michiel Marcus. "Privacy-Preserving Contrastive Explanations with Local Foil Trees." Cryptography 6, no. 4 (October 28, 2022): 54. http://dx.doi.org/10.3390/cryptography6040054.
Full textAbstract:
We present the first algorithm that combines privacy-preserving technologies and state-of-the-art explainable AI to enable privacy-friendly explanations of black-box AI models. We provide a secure algorithm for contrastive explanations of black-box machine learning models that securely trains and uses local foil trees. Our work shows that the quality of these explanations can be upheld whilst ensuring the privacy of both the training data and the model itself.
11
Pulatov, Damir, and Lars Kotthoff. "Opening the Black Box: Automatically Characterizing Software for Algorithm Selection (Student Abstract)." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 10 (April 3, 2020): 13899–900. http://dx.doi.org/10.1609/aaai.v34i10.7222.
Full textAbstract:
Meta-algorithmics, the field of leveraging machine learning to use algorithms more efficiently, has achieved impressive performance improvements in many areas of AI. It treats the algorithms to improve on as black boxes – nothing is known about their inner workings. This allows meta-algorithmic techniques to be deployed in many applications, but leaves potential performance improvements untapped by ignoring information that the algorithms could provide. In this paper, we open the black box without sacrificing the universal applicability of meta-algorithmic techniques by automatically analyzing the source code of the algorithms under consideration and show how to use it to improve algorithm selection performance. We demonstrate improvements of up to 82% on the standard ASlib benchmark library.
12
BALL, NICHOLAS M., and ROBERT J. BRUNNER. "DATA MINING AND MACHINE LEARNING IN ASTRONOMY." International Journal of Modern Physics D 19, no. 07 (July 2010): 1049–106. http://dx.doi.org/10.1142/s0218271810017160.
Full textAbstract:
We review the current state of data mining and machine learning in astronomy. Data Mining can have a somewhat mixed connotation from the point of view of a researcher in this field. If used correctly, it can be a powerful approach, holding the potential to fully exploit the exponentially increasing amount of available data, promising great scientific advance. However, if misused, it can be little more than the black box application of complex computing algorithms that may give little physical insight, and provide questionable results. Here, we give an overview of the entire data mining process, from data collection through to the interpretation of results. We cover common machine learning algorithms, such as artificial neural networks and support vector machines, applications from a broad range of astronomy, emphasizing those in which data mining techniques directly contributed to improving science, and important current and future directions, including probability density functions, parallel algorithms, Peta-Scale computing, and the time domain. We conclude that, so long as one carefully selects an appropriate algorithm and is guided by the astronomical problem at hand, data mining can be very much the powerful tool, and not the questionable black box.
13
Yu, Wen, and Francisco Vega. "Nonlinear system modeling using the takagi-sugeno fuzzy model and long-short term memory cells." Journal of Intelligent & Fuzzy Systems 39, no. 3 (October 7, 2020): 4547–56. http://dx.doi.org/10.3233/jifs-200491.
Full textAbstract:
The data driven black-box or gray-box models like neural networks and fuzzy systems have some disadvantages, such as the high and uncertain dimensions and complex learning process. In this paper, we combine the Takagi-Sugeno fuzzy model with long-short term memory cells to overcome these disadvantages. This novel model takes the advantages of the interpretability of the fuzzy system and the good approximation ability of the long-short term memory cell. We propose a fast and stable learning algorithm for this model. Comparisons with others similar black-box and grey-box models are made, in order to observe the advantages of the proposal.
14
Muñoz, Mario Andrés, and Michael Kirley. "Sampling Effects on Algorithm Selection for Continuous Black-Box Optimization." Algorithms 14, no. 1 (January 11, 2021): 19. http://dx.doi.org/10.3390/a14010019.
Full textAbstract:
In this paper, we investigate how systemic errors due to random sampling impact on automated algorithm selection for bound-constrained, single-objective, continuous black-box optimization. We construct a machine learning-based algorithm selector, which uses exploratory landscape analysis features as inputs. We test the accuracy of the recommendations experimentally using resampling techniques and the hold-one-instance-out and hold-one-problem-out validation methods. The results demonstrate that the selector remains accurate even with sampling noise, although not without trade-offs.
15
Muñoz, Mario Andrés, and Michael Kirley. "Sampling Effects on Algorithm Selection for Continuous Black-Box Optimization." Algorithms 14, no. 1 (January 11, 2021): 19. http://dx.doi.org/10.3390/a14010019.
Full textAbstract:
In this paper, we investigate how systemic errors due to random sampling impact on automated algorithm selection for bound-constrained, single-objective, continuous black-box optimization. We construct a machine learning-based algorithm selector, which uses exploratory landscape analysis features as inputs. We test the accuracy of the recommendations experimentally using resampling techniques and the hold-one-instance-out and hold-one-problem-out validation methods. The results demonstrate that the selector remains accurate even with sampling noise, although not without trade-offs.
16
Žlahtič, Bojan, Jernej Završnik, Helena Blažun Vošner, Peter Kokol, David Šuran, and Tadej Završnik. "Agile Machine Learning Model Development Using Data Canyons in Medicine: A Step towards Explainable Artificial Intelligence and Flexible Expert-Based Model Improvement." Applied Sciences 13, no. 14 (July 19, 2023): 8329. http://dx.doi.org/10.3390/app13148329.
Full textAbstract:
Over the past few decades, machine learning has emerged as a valuable tool in the field of medicine, driven by the accumulation of vast amounts of medical data and the imperative to harness this data for the betterment of humanity. However, many of the prevailing machine learning algorithms in use today are characterized as black-box models, lacking transparency in their decision-making processes and are often devoid of clear visualization capabilities. The transparency of these machine learning models impedes medical experts from effectively leveraging them due to the high-stakes nature of their decisions. Consequently, the need for explainable artificial intelligence (XAI) that aims to address the demand for transparency in the decision-making mechanisms of black-box algorithms has arisen. Alternatively, employing white-box algorithms can empower medical experts by allowing them to contribute their knowledge to the decision-making process and obtain a clear and transparent output. This approach offers an opportunity to personalize machine learning models through an agile process. A novel white-box machine learning algorithm known as Data canyons was employed as a transparent and robust foundation for the proposed solution. By providing medical experts with a web framework where their expertise is transferred to a machine learning model and enabling the utilization of this process in an agile manner, a symbiotic relationship is fostered between the domains of medical expertise and machine learning. The flexibility to manipulate the output machine learning model and visually validate it, even without expertise in machine learning, establishes a crucial link between these two expert domains.
17
HOLZINGER, ANDREAS, MARKUS PLASS, KATHARINA HOLZINGER, GLORIA CERASELA CRIS¸AN, CAMELIA-M. PINTEA, and VASILE PALADE. "A glass-box interactive machine learning approach for solving NP-hard problems with the human-in-the-loop." Creative Mathematics and Informatics 28, no. 2 (June 20, 2019): 121–34. http://dx.doi.org/10.37193/cmi.2019.02.04.
Full textAbstract:
The ultimate goal of the Machine Learning (ML) community is to develop algorithms that can automatically learn from data, to extract knowledge and to make decisions without any human intervention. Specifically, automatic Machine Learning (aML) approaches show impressive success, e.g. in speech/image recognition or autonomous drive and smart car industry. Recent results even demonstrate intriguingly that deep learning applied for automatic classification of skin lesions is on par with the performance of dermatologists, yet outperforms the average human efficiency. As human perception is inherently limited to 3D environments, such approaches can discover patterns, e.g. that two objects are similar, in arbitrarily high-dimensional spaces what no human is able to do. Humans can deal simultaneously only with limited amounts of data, whilst “big data” is not only beneficial but necessary for aML. However, in health informatics, there are few data sets; aML approaches often suffer from insufficient training samples. Many problems are computationally hard, e.g. subspace clustering, k-anonymization, or protein folding. Here, interactive machine learning (iML) could be successfully used, as a human-in-the-loop contributes to reduce a huge search space through heuristic selection of suitable samples. This can reduce the complexity of NP-hard problems through the knowledge brought in by a human agent involved into the learning algorithm. A huge motivation for iML is that standard black-box approaches lack transparency, hence do not foster trust and acceptance of ML among end-users. Most of all, rising legal and privacy aspects, e.g. the European General Data Protection Regulations (GDPR) make black-box approaches difficult to use, because they often are not able to explain why a decision has been made, e.g. why two objects are similar. All these reasons motivate the idea to open the black-box to a glass-box. In this paper, we present some experiments to demonstrate the effectiveness of the iML human-in-the-loop model, in particular when using a glass-box instead of a black-box model and thus enabling a human directly to interact with a learning algorithm. We selected the Ant Colony System (ACS) algorithm, and applied it on the Traveling Salesman Problem (TSP). The TSP-problem is a good example, because it is of high relevance for health informatics as for example on protein folding problem, thus of enormous importance for fostering cancer research. Finally, from studies of learning from observation, i.e. of how humans extract so much from so little data, fundamental ML-research also may benefit.
18
Saokani, Ukan, Mohamad Irfan, Dian Sa'adillah Maylawati, Rachmat Jaenal Abidin, Ichsan Taufik, and Riyan Naufal Hay's. "Comparison of the Fisher-Yates Shuffle and the Linear Congruent Algorithm for Randomizing Questions in Nahwu Learning Multimedia." Khazanah Journal of Religion and Technology 1, no. 1 (June 1, 2023): 10–14. http://dx.doi.org/10.15575/kjrt.v1i1.159.
Full textAbstract:
Nahwu Quiz is a basic Arabic learning application that can be played by the public over the age of 12 years. In the question practice menu, there are questions and 4 multiple choice questions. The user only needs to choose one of the multiple choices that the user thinks is correct/matches the question at hand. In one game, there are 5 questions. After answering all these questions, you will immediately see the score. The purpose of developing this application apart from being a medium of entertainment as well as a medium of learning and memory training for game users (users). To make this Nahwu Quiz application, the authors use the Fisher Yates Shuffle (FYS) algorithm which is used to perform the randomization function in multiple choice and the Linear Congruent Method (LCM) algorithm as a comparison. White box and black box testing were applied to see the feasibility of the program and to obtain efficiency in the comparison of randomization methods. The results of white box and black box testing on the application show that the application is feasible. with reference to the white box test results that the FYS algorithm and the LCM have the same complexity as the result of cyclomatic complexity = 2.
19
Wongvibulsin, Shannon, Katherine C. Wu, and Scott L. Zeger. "Improving Clinical Translation of Machine Learning Approaches Through Clinician-Tailored Visual Displays of Black Box Algorithms: Development and Validation." JMIR Medical Informatics 8, no. 6 (June 9, 2020): e15791. http://dx.doi.org/10.2196/15791.
Full textAbstract:
Background Despite the promise of machine learning (ML) to inform individualized medical care, the clinical utility of ML in medicine has been limited by the minimal interpretability and black box nature of these algorithms. Objective The study aimed to demonstrate a general and simple framework for generating clinically relevant and interpretable visualizations of black box predictions to aid in the clinical translation of ML. Methods To obtain improved transparency of ML, simplified models and visual displays can be generated using common methods from clinical practice such as decision trees and effect plots. We illustrated the approach based on postprocessing of ML predictions, in this case random forest predictions, and applied the method to data from the Left Ventricular (LV) Structural Predictors of Sudden Cardiac Death (SCD) Registry for individualized risk prediction of SCD, a leading cause of death. Results With the LV Structural Predictors of SCD Registry data, SCD risk predictions are obtained from a random forest algorithm that identifies the most important predictors, nonlinearities, and interactions among a large number of variables while naturally accounting for missing data. The black box predictions are postprocessed using classification and regression trees into a clinically relevant and interpretable visualization. The method also quantifies the relative importance of an individual or a combination of predictors. Several risk factors (heart failure hospitalization, cardiac magnetic resonance imaging indices, and serum concentration of systemic inflammation) can be clearly visualized as branch points of a decision tree to discriminate between low-, intermediate-, and high-risk patients. Conclusions Through a clinically important example, we illustrate a general and simple approach to increase the clinical translation of ML through clinician-tailored visual displays of results from black box algorithms. We illustrate this general model-agnostic framework by applying it to SCD risk prediction. Although we illustrate the methods using SCD prediction with random forest, the methods presented are applicable more broadly to improving the clinical translation of ML, regardless of the specific ML algorithm or clinical application. As any trained predictive model can be summarized in this manner to a prespecified level of precision, we encourage the use of simplified visual displays as an adjunct to the complex predictive model. Overall, this framework can allow clinicians to peek inside the black box and develop a deeper understanding of the most important features from a model to gain trust in the predictions and confidence in applying them to clinical care.
20
Lu, Li, Yizhong Wu, Qi Zhang, and Ping Qiao. "A Transformation-Based Improved Kriging Method for the Black Box Problem in Reliability-Based Design Optimization." Mathematics 11, no. 1 (January 1, 2023): 218. http://dx.doi.org/10.3390/math11010218.
Full textAbstract:
In order to overcome the drawbacks of expensive function evaluation in the practical reliability-based design optimization (RBDO) problem, researchers have proposed the black box-based RBDO method. The algorithm flow of the commonly employed RBDO method for the black box problem consists of the outer construction loop of the surrogate model of the constraint function and the inner surrogate model-based solving loop. To improve the solving ability of the black box RBDO problem, this paper proposes a transformation-based improved kriging method to increase the effectiveness of the two loops identified above. For the outer loop, a sample distribution-based learning function is suggested to improve the construction efficiency of the surrogate model of the constraint function. For the inner loop, a paired incremental sample-based limit reliability boundary construction approach is suggested to transform the RBDO problem into an equivalent deterministic design optimization problem that can be efficiently solved by classical optimization algorithms. The test results of five cases demonstrate that the proposed method can accurately construct the surrogate model of the constraint function and efficiently solve the black box RBDO problem.
21
Kerschke, Pascal, and Heike Trautmann. "Automated Algorithm Selection on Continuous Black-Box Problems by Combining Exploratory Landscape Analysis and Machine Learning." Evolutionary Computation 27, no. 1 (March 2019): 99–127. http://dx.doi.org/10.1162/evco_a_00236.
Full textAbstract:
In this article, we build upon previous work on designing informative and efficient Exploratory Landscape Analysis features for characterizing problems' landscapes and show their effectiveness in automatically constructing algorithm selection models in continuous black-box optimization problems. Focusing on algorithm performance results of the COCO platform of several years, we construct a representative set of high-performing complementary solvers and present an algorithm selection model that, compared to the portfolio's single best solver, on average requires less than half of the resources for solving a given problem. Therefore, there is a huge gain in efficiency compared to classical ensemble methods combined with an increased insight into problem characteristics and algorithm properties by using informative features. The model acts on the assumption that the function set of the Black-Box Optimization Benchmark is representative enough for practical applications. The model allows for selecting the best suited optimization algorithm within the considered set for unseen problems prior to the optimization itself based on a small sample of function evaluations. Note that such a sample can even be reused for the initial population of an evolutionary (optimization) algorithm so that even the feature costs become negligible.
22
Possatto, André Bina. "Painting the black box white: Interpreting an algorithm-based trading strategy." Brazilian Review of Finance 20, no. 3 (September 11, 2022): 105–38. http://dx.doi.org/10.12660/rbfin.v20n3.2022.81999.
Full textAbstract:
Difficulty understanding how a black box model makes predictions undermines machine learning's success in financial markets. We show how to employ model-agnostic methods to carry out machine learning stock market predictions that are more transparent to a human investor. We create long-short investment strategies using a tree-based fundamental analysis. We apply the models to the Brazilian stock market, achieving an out-of-sample expected annual return of 26.4% with a Sharpe ratio of 0.50. Ensembles between the long and short legs improve Sharpe ratio up to 1.26. Our strategy has low asset turnover and hence transaction costs do not harm performance too much. Interpretation shows differences in the main drivers of over- and underperformance.
23
Verma, Pulkit, Shashank Rao Marpally, and Siddharth Srivastava. "Asking the Right Questions: Learning Interpretable Action Models Through Query Answering." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 13 (May 18, 2021): 12024–33. http://dx.doi.org/10.1609/aaai.v35i13.17428.
Full textAbstract:
This paper develops a new approach for estimating an interpretable, relational model of a black-box autonomous agent that can plan and act. Our main contributions are a new paradigm for estimating such models using a rudimentary query interface with the agent and a hierarchical querying algorithm that generates an interrogation policy for estimating the agent's internal model in a user-interpretable vocabulary. Empirical evaluation of our approach shows that despite the intractable search space of possible agent models, our approach allows correct and scalable estimation of interpretable agent models for a wide class of black-box autonomous agents. Our results also show that this approach can use predicate classifiers to learn interpretable models of planning agents that represent states as images.
24
Zhu, Mingzhe, Jie Cheng, Tao Lei, Zhenpeng Feng, Xianda Zhou, Yuanjing Liu, and Zhihan Chen. "C-RISE: A Post-Hoc Interpretation Method of Black-Box Models for SAR ATR." Remote Sensing 15, no. 12 (June 14, 2023): 3103. http://dx.doi.org/10.3390/rs15123103.
Full textAbstract:
The integration of deep learning methods, especially Convolutional Neural Networks (CNN), and Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) has been widely deployed in the field of radar signal processing. Nevertheless, these methods are frequently regarded as black-box models due to the limited visual interpretation of their internal feature representation and parameter organization. In this paper, we propose an innovative approach named C-RISE, which builds upon the RISE algorithm to provide a post-hoc interpretation technique for black-box models used in SAR Images Target Recognition. C-RISE generates saliency maps that effectively visualize the significance of each pixel. Our algorithm outperforms RISE by clustering masks that capture similar fusion features into distinct groups, enabling more appropriate weight distribution and increased focus on the target area. Furthermore, we employ Gaussian blur to process the masked area, preserving the original image structure with optimal consistency and integrity. C-RISE has been extensively evaluated through experiments, and the results demonstrate superior performance over other interpretation methods based on perturbation when applied to neural networks for SAR image target recognition. Furthermore, our approach is highly robust and transferable compared to other interpretable algorithms, including white-box methods.
25
Sudry, Matan, and Erez Karpas. "Learning to Estimate Search Progress Using Sequence of States." Proceedings of the International Conference on Automated Planning and Scheduling 32 (June 13, 2022): 362–70. http://dx.doi.org/10.1609/icaps.v32i1.19821.
Full textAbstract:
Many problems of interest can be solved using heuristic search algorithms. When solving a heuristic search problem, we are often interested in estimating search progress, that is, how much longer until we have a solution. Previous work on search progress estimation derived formulas based on some relevant features that can be observed from the behavior of the search algorithm. In this paper, rather than manually deriving such formulas we leverage machine learning to learn more accurate search progress predictors automatically. We train a Long Short-Term Memory (LSTM) network, which takes as input sequences of nodes expanded by the search algorithm, and predicts how far along with the search we are. Importantly, our approach still treats the search algorithm as a black box and does not look into the contents of search nodes. An empirical evaluation shows our technique outperforms previous search progress estimation techniques.
26
Englert, Peter, and Marc Toussaint. "Learning manipulation skills from a single demonstration." International Journal of Robotics Research 37, no. 1 (December 5, 2017): 137–54. http://dx.doi.org/10.1177/0278364917743795.
Full textAbstract:
We consider the scenario where a robot is demonstrated a manipulation skill once and should then use only a few trials on its own to learn to reproduce, optimize, and generalize that same skill. A manipulation skill is generally a high-dimensional policy. To achieve the desired sample efficiency, we need to exploit the inherent structure in this problem. With our approach, we propose to decompose the problem into analytically known objectives, such as motion smoothness, and black-box objectives, such as trial success or reward, depending on the interaction with the environment. The decomposition allows us to leverage and combine (i) constrained optimization methods to address analytic objectives, (ii) constrained Bayesian optimization to explore black-box objectives, and (iii) inverse optimal control methods to eventually extract a generalizable skill representation. The algorithm is evaluated on a synthetic benchmark experiment and compared with state-of-the-art learning methods. We also demonstrate the performance on real-robot experiments with a PR2.
27
Yuan, Mu, Lan Zhang, and Xiang-Yang Li. "MLink: Linking Black-Box Models for Collaborative Multi-Model Inference." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 9 (June 28, 2022): 9475–83. http://dx.doi.org/10.1609/aaai.v36i9.21180.
Full textAbstract:
The cost efficiency of model inference is critical to real-world machine learning (ML) applications, especially for delay-sensitive tasks and resource-limited devices. A typical dilemma is: in order to provide complex intelligent services (e.g. smart city), we need inference results of multiple ML models, but the cost budget (e.g. GPU memory) is not enough to run all of them. In this work, we study underlying relationships among black-box ML models and propose a novel learning task: model linking. Model linking aims to bridge the knowledge of different black-box models by learning mappings (dubbed model links) between their output spaces. Based on model links, we developed a scheduling algorithm, named MLink. Through collaborative multi-model inference enabled by model links, MLink can improve the accuracy of obtained inference results under the cost budget. We evaluated MLink on a multi-modal dataset with seven different ML models and two real-world video analytics systems with six ML models and 3,264 hours of video. Experimental results show that our proposed model links can be effectively built among various black-box models. Under the budget of GPU memory, MLink can save 66.7% inference computations while preserving 94% inference accuracy, which outperforms multi-task learning, deep reinforcement learning-based scheduler and frame filtering baselines.
28
Wang, Fangwei, Yuanyuan Lu, Changguang Wang, and Qingru Li. "Binary Black-Box Adversarial Attacks with Evolutionary Learning against IoT Malware Detection." Wireless Communications and Mobile Computing 2021 (August 30, 2021): 1–9. http://dx.doi.org/10.1155/2021/8736946.
Full textAbstract:
5G is about to open Pandora’s box of security threats to the Internet of Things (IoT). Key technologies, such as network function virtualization and edge computing introduced by the 5G network, bring new security threats and risks to the Internet infrastructure. Therefore, higher detection and defense against malware are required. Nowadays, deep learning (DL) is widely used in malware detection. Recently, research has demonstrated that adversarial attacks have posed a hazard to DL-based models. The key issue of enhancing the antiattack performance of malware detection systems that are used to detect adversarial attacks is to generate effective adversarial samples. However, numerous existing methods to generate adversarial samples are manual feature extraction or using white-box models, which makes it not applicable in the actual scenarios. This paper presents an effective binary manipulation-based attack framework, which generates adversarial samples with an evolutionary learning algorithm. The framework chooses some appropriate action sequences to modify malicious samples. Thus, the modified malware can successfully circumvent the detection system. The evolutionary algorithm can adaptively simplify the modification actions and make the adversarial sample more targeted. Our approach can efficiently generate adversarial samples without human intervention. The generated adversarial samples can effectively combat DL-based malware detection models while preserving the consistency of the executable and malicious behavior of the original malware samples. We apply the generated adversarial samples to attack the detection engines of VirusTotal. Experimental results illustrate that the adversarial samples generated by our method reach an evasion success rate of 47.8%, which outperforms other attack methods. By adding adversarial samples in the training process, the MalConv network is retrained. We show that the detection accuracy is improved by 10.3%.
29
Cretu, Andrei. "Learning the Ashby Box: an experiment in second order cybernetic modeling." Kybernetes 49, no. 8 (November 23, 2019): 2073–90. http://dx.doi.org/10.1108/k-06-2019-0439.
Full textAbstract:
Purpose W. Ross Ashby’s elementary non-trivial machine, known in the cybernetic literature as the “Ashby Box,” has been described as the prototypical example of a black box system. As far as it can be ascertained from Ashby’s journal, the intended purpose of this device may have been to exemplify the environment where an “artificial brain” may operate. This paper describes the construction of an elementary observer/controller for the class of systems exemplified by the Ashby Box – variable structure black box systems with parallel input. Design/methodology/approach Starting from a formalization of the second-order assumptions implicit in the design of the Ashby Box, the observer/controller system is synthesized from the ground up, in a strictly system-theoretic setting, without recourse to disciplinary metaphors or current theories of learning and cognition, based mainly on guidance from Heinz von Foerster’s theory of self-organizing systems and W. Ross Ashby’s own insights into adaptive systems. Findings Achieving and maintaining control of the Ashby Box requires a non-trivial observer system able to use the results of its interactions with the non-trivial machine to autonomously construct, deconstruct and reconstruct its own function. The algorithm and the dynamical model of the Ashby Box observer developed in this paper define the basic specifications of a general purpose, unsupervised learning architecture able to accomplish this task. Originality/value The problem exemplified by the Ashby Box is fundamental and goes to the roots of cybernetic theory; second-order cybernetics offers an adequate foundation for the mathematical modeling of this problem.
30
Li, Zun, and Michael Wellman. "Structure Learning for Approximate Solution of Many-Player Games." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 02 (April 3, 2020): 2119–27. http://dx.doi.org/10.1609/aaai.v34i02.5586.
Full textAbstract:
Games with many players are difficult to solve or even specify without adopting structural assumptions that enable representation in compact form. Such structure is generally not given and will not hold exactly for particular games of interest. We introduce an iterative structure-learning approach to search for approximate solutions of many-player games, assuming only black-box simulation access to noisy payoff samples. Our first algorithm, K-Roles, exploits symmetry by learning a role assignment for players of the game through unsupervised learning (clustering) methods. Our second algorithm, G3L, seeks sparsity by greedy search over local interactions to learn a graphical game model. Both algorithms use supervised learning (regression) to fit payoff values to the learned structures, in compact representations that facilitate equilibrium calculation. We experimentally demonstrate the efficacy of both methods in reaching quality solutions and uncovering hidden structure, on both perfectly and approximately structured game instances.
31
Shahpouri, Saeid, Armin Norouzi, Christopher Hayduk, Reza Rezaei, Mahdi Shahbakhti, and Charles Robert Koch. "Hybrid Machine Learning Approaches and a Systematic Model Selection Process for Predicting Soot Emissions in Compression Ignition Engines." Energies 14, no. 23 (November 24, 2021): 7865. http://dx.doi.org/10.3390/en14237865.
Full textAbstract:
The standards for emissions from diesel engines are becoming more stringent and accurate emission modeling is crucial in order to control the engine to meet these standards. Soot emissions are formed through a complex process and are challenging to model. A comprehensive analysis of diesel engine soot emissions modeling for control applications is presented in this paper. Physical, black-box, and gray-box models are developed for soot emissions prediction. Additionally, different feature sets based on the least absolute shrinkage and selection operator (LASSO) feature selection method and physical knowledge are examined to develop computationally efficient soot models with good precision. The physical model is a virtual engine modeled in GT-Power software that is parameterized using a portion of experimental data. Different machine learning methods, including Regression Tree (RT), Ensemble of Regression Trees (ERT), Support Vector Machines (SVM), Gaussian Process Regression (GPR), Artificial Neural Network (ANN), and Bayesian Neural Network (BNN) are used to develop the black-box models. The gray-box models include a combination of the physical and black-box models. A total of five feature sets and eight different machine learning methods are tested. An analysis of the accuracy, training time and test time of the models is performed using the K-means clustering algorithm. It provides a systematic way for categorizing the feature sets and methods based on their performance and selecting the best method for a specific application. According to the analysis, the black-box model consisting of GPR and feature selection by LASSO shows the best performance with test R2 of 0.96. The best gray-box model consists of SVM-based method with physical insight feature set along with LASSO for feature selection with test R2 of 0.97.
32
Bizzo, Bernardo C., Shadi Ebrahimian, Mark E. Walters, Mark H. Michalski, Katherine P. Andriole, Keith J. Dreyer, Mannudeep K. Kalra, Tarik Alkasab, and Subba R. Digumarthy. "Validation pipeline for machine learning algorithm assessment for multiple vendors." PLOS ONE 17, no. 4 (April 29, 2022): e0267213. http://dx.doi.org/10.1371/journal.pone.0267213.
Full textAbstract:
A standardized objective evaluation method is needed to compare machine learning (ML) algorithms as these tools become available for clinical use. Therefore, we designed, built, and tested an evaluation pipeline with the goal of normalizing performance measurement of independently developed algorithms, using a common test dataset of our clinical imaging. Three vendor applications for detecting solid, part-solid, and groundglass lung nodules in chest CT examinations were assessed in this retrospective study using our data-preprocessing and algorithm assessment chain. The pipeline included tools for image cohort creation and de-identification; report and image annotation for ground-truth labeling; server partitioning to receive vendor “black box” algorithms and to enable model testing on our internal clinical data (100 chest CTs with 243 nodules) from within our security firewall; model validation and result visualization; and performance assessment calculating algorithm recall, precision, and receiver operating characteristic curves (ROC). Algorithm true positives, false positives, false negatives, recall, and precision for detecting lung nodules were as follows: Vendor-1 (194, 23, 49, 0.80, 0.89); Vendor-2 (182, 270, 61, 0.75, 0.40); Vendor-3 (75, 120, 168, 0.32, 0.39). The AUCs for detection of solid (0.61–0.74), groundglass (0.66–0.86) and part-solid (0.52–0.86) nodules varied between the three vendors. Our ML model validation pipeline enabled testing of multi-vendor algorithms within the institutional firewall. Wide variations in algorithm performance for detection as well as classification of lung nodules justifies the premise for a standardized objective ML algorithm evaluation process.
33
McTavish, Hayden, Chudi Zhong, Reto Achermann, Ilias Karimalis, Jacques Chen, Cynthia Rudin, and Margo Seltzer. "Fast Sparse Decision Tree Optimization via Reference Ensembles." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 9 (June 28, 2022): 9604–13. http://dx.doi.org/10.1609/aaai.v36i9.21194.
Full textAbstract:
Sparse decision tree optimization has been one of the most fundamental problems in AI since its inception and is a challenge at the core of interpretable machine learning. Sparse decision tree optimization is computationally hard, and despite steady effort since the 1960's, breakthroughs have been made on the problem only within the past few years, primarily on the problem of finding optimal sparse decision trees. However, current state-of-the-art algorithms often require impractical amounts of computation time and memory to find optimal or near-optimal trees for some real-world datasets, particularly those having several continuous-valued features. Given that the search spaces of these decision tree optimization problems are massive, can we practically hope to find a sparse decision tree that competes in accuracy with a black box machine learning model? We address this problem via smart guessing strategies that can be applied to any optimal branch-and-bound-based decision tree algorithm. The guesses come from knowledge gleaned from black box models. We show that by using these guesses, we can reduce the run time by multiple orders of magnitude while providing bounds on how far the resulting trees can deviate from the black box's accuracy and expressive power. Our approach enables guesses about how to bin continuous features, the size of the tree, and lower bounds on the error for the optimal decision tree. Our experiments show that in many cases we can rapidly construct sparse decision trees that match the accuracy of black box models. To summarize: when you are having trouble optimizing, just guess.
34
Van Calster, Ben, Laure Wynants, Dirk Timmerman, Ewout W. Steyerberg, and Gary S. Collins. "Predictive analytics in health care: how can we know it works?" Journal of the American Medical Informatics Association 26, no. 12 (August 2, 2019): 1651–54. http://dx.doi.org/10.1093/jamia/ocz130.
Full textAbstract:
Abstract There is increasing awareness that the methodology and findings of research should be transparent. This includes studies using artificial intelligence to develop predictive algorithms that make individualized diagnostic or prognostic risk predictions. We argue that it is paramount to make the algorithm behind any prediction publicly available. This allows independent external validation, assessment of performance heterogeneity across settings and over time, and algorithm refinement or updating. Online calculators and apps may aid uptake if accompanied with sufficient information. For algorithms based on “black box” machine learning methods, software for algorithm implementation is a must. Hiding algorithms for commercial exploitation is unethical, because there is no possibility to assess whether algorithms work as advertised or to monitor when and how algorithms are updated. Journals and funders should demand maximal transparency for publications on predictive algorithms, and clinical guidelines should only recommend publicly available algorithms.
35
Yin, Yiqiao, and Yash Bingi. "Using Machine Learning to Classify Human Fetal Health and Analyze Feature Importance." BioMedInformatics 3, no. 2 (April 1, 2023): 280–98. http://dx.doi.org/10.3390/biomedinformatics3020019.
Full textAbstract:
The reduction of childhood mortality is an ongoing struggle and a commonly used factor in determining progress in the medical field. The under-5 mortality number is around 5 million around the world, with many of the deaths being preventable. In light of this issue, cardiotocograms (CTGs) have emerged as a leading tool to determine fetal health. By using ultrasound pulses and reading the responses, CTGs help healthcare professionals assess the overall health of the fetus to determine the risk of child mortality. However, interpreting the results of the CTGs is time consuming and inefficient, especially in underdeveloped areas where an expert obstetrician is hard to come by. Using a support vector machine (SVM) and oversampling, this paper proposes a model that classifies fetal health with an accuracy of 99.59%. To further explain the CTG measurements, an algorithm based off of RISE (Randomized Input Sampling for Explanation of Black-box Models) was created, called Feature Alteration for explanation of Black Box Models (FAB). The findings of this novel algorithm were compared to SHapley Additive exPlanations (SHAP) and Local Interpretable Model Agnostic Explanations (LIME). Overall, this technology allows doctors and medical professionals to classify fetal health with high accuracy and determine which features were most influential in the process.
36
Aslam, Nida, Irfan Ullah Khan, Samiha Mirza, Alanoud AlOwayed, Fatima M. Anis, Reef M. Aljuaid, and Reham Baageel. "Interpretable Machine Learning Models for Malicious Domains Detection Using Explainable Artificial Intelligence (XAI)." Sustainability 14, no. 12 (June 16, 2022): 7375. http://dx.doi.org/10.3390/su14127375.
Full textAbstract:
With the expansion of the internet, a major threat has emerged involving the spread of malicious domains intended by attackers to perform illegal activities aiming to target governments, violating privacy of organizations, and even manipulating everyday users. Therefore, detecting these harmful domains is necessary to combat the growing network attacks. Machine Learning (ML) models have shown significant outcomes towards the detection of malicious domains. However, the “black box” nature of the complex ML models obstructs their wide-ranging acceptance in some of the fields. The emergence of Explainable Artificial Intelligence (XAI) has successfully incorporated the interpretability and explicability in the complex models. Furthermore, the post hoc XAI model has enabled the interpretability without affecting the performance of the models. This study aimed to propose an Explainable Artificial Intelligence (XAI) model to detect malicious domains on a recent dataset containing 45,000 samples of malicious and non-malicious domains. In the current study, initially several interpretable ML models, such as Decision Tree (DT) and Naïve Bayes (NB), and black box ensemble models, such as Random Forest (RF), Extreme Gradient Boosting (XGB), AdaBoost (AB), and Cat Boost (CB) algorithms, were implemented and found that XGB outperformed the other classifiers. Furthermore, the post hoc XAI global surrogate model (Shapley additive explanations) and local surrogate LIME were used to generate the explanation of the XGB prediction. Two sets of experiments were performed; initially the model was executed using a preprocessed dataset and later with selected features using the Sequential Forward Feature selection algorithm. The results demonstrate that ML algorithms were able to distinguish benign and malicious domains with overall accuracy ranging from 0.8479 to 0.9856. The ensemble classifier XGB achieved the highest result, with an AUC and accuracy of 0.9991 and 0.9856, respectively, before the feature selection algorithm, while there was an AUC of 0.999 and accuracy of 0.9818 after the feature selection algorithm. The proposed model outperformed the benchmark study.
37
Soucha, Michal, and Kirill Bogdanov. "Observation Tree Approach: Active Learning Relying on Testing." Computer Journal 63, no. 9 (July 3, 2019): 1298–310. http://dx.doi.org/10.1093/comjnl/bxz056.
Full textAbstract:
Abstract The correspondence of active learning and testing of finite-state machines (FSMs) has been known for a while; however, it was not utilized in the learning. We propose a new framework called the observation tree approach that allows one to use the testing theory to improve the performance of active learning. The improvement is demonstrated on three novel learning algorithms that implement the observation tree approach. They outperform the standard learning algorithms, such as the L* algorithm, in the setting where a minimally adequate teacher provides counterexamples. Moreover, they can also significantly reduce the dependency on the teacher using the assumption of extra states that is well-established in the testing of FSMs. This is immensely helpful as a teacher does not have to be available if one learns a model of a black box, such as a system only accessible via a network.
38
Patil, Vishakha, Ganesh Ghalme, Vineet Nair, and Y. Narahari. "Achieving Fairness in the Stochastic Multi-Armed Bandit Problem." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 04 (April 3, 2020): 5379–86. http://dx.doi.org/10.1609/aaai.v34i04.5986.
Full textAbstract:
We study an interesting variant of the stochastic multi-armed bandit problem, which we call the Fair-MAB problem, where, in addition to the objective of maximizing the sum of expected rewards, the algorithm also needs to ensure that at any time, each arm is pulled at least a pre-specified fraction of times. We investigate the interplay between learning and fairness in terms of a pre-specified vector denoting the fractions of guaranteed pulls. We define a fairness-aware regret, which we call r-Regret, that takes into account the above fairness constraints and extends the conventional notion of regret in a natural way. Our primary contribution is to obtain a complete characterization of a class of Fair-MAB algorithms via two parameters: the unfairness tolerance and the learning algorithm used as a black-box. For this class of algorithms, we provide a fairness guarantee that holds uniformly over time, irrespective of the choice of the learning algorithm. Further, when the learning algorithm is UCB1, we show that our algorithm achieves constant r-Regret for a large enough time horizon. Finally, we analyze the cost of fairness in terms of the conventional notion of regret. We conclude by experimentally validating our theoretical results.
39
Li, Yuancheng, and Yimeng Wang. "Defense Against Adversarial Attacks in Deep Learning." Applied Sciences 9, no. 1 (December 26, 2018): 76. http://dx.doi.org/10.3390/app9010076.
Full textAbstract:
Neural networks are very vulnerable to adversarial examples, which threaten their application in security systems, such as face recognition, and autopilot. In response to this problem, we propose a new defensive strategy. In our strategy, we propose a new deep denoising neural network, which is called UDDN, to remove the noise on adversarial samples. The standard denoiser suffers from the amplification effect, in which the small residual adversarial noise gradually increases and leads to misclassification. The proposed denoiser overcomes this problem by using a special loss function, which is defined as the difference between the model outputs activated by the original image and denoised image. At the same time, we propose a new model training algorithm based on knowledge transfer, which can resist slight image disturbance and make the model generalize better around the training samples. Our proposed defensive strategy is robust against both white-box or black-box attacks. Meanwhile, the strategy is applicable to any deep neural network-based model. In the experiment, we apply the defensive strategy to a face recognition model. The experimental results show that our algorithm can effectively resist adversarial attacks and improve the accuracy of the model.
40
Samaras, Agorastos-Dimitrios, Serafeim Moustakidis, Ioannis D. Apostolopoulos, Elpiniki Papageorgiou, and Nikolaos Papandrianos. "Uncovering the Black Box of Coronary Artery Disease Diagnosis: The Significance of Explainability in Predictive Models." Applied Sciences 13, no. 14 (July 12, 2023): 8120. http://dx.doi.org/10.3390/app13148120.
Full textAbstract:
In recent times, coronary artery disease (CAD) prediction and diagnosis have been the subject of many Medical decision support systems (MDSS) that make use of machine learning (ML) and deep learning (DL) algorithms. The common ground of most of these applications is that they function as black boxes. They reach a conclusion/diagnosis using multiple features as input; however, the user is oftentimes oblivious to the prediction process and the feature weights leading to the eventual prediction. The primary objective of this study is to enhance the transparency and comprehensibility of a black-box prediction model designed for CAD. The dataset employed in this research comprises biometric and clinical information obtained from 571 patients, encompassing 21 different features. Among the instances, 43% of cases of CAD were confirmed through invasive coronary angiography (ICA). Furthermore, a prediction model utilizing the aforementioned dataset and the CatBoost algorithm is analyzed to highlight its prediction making process and the significance of each input datum. State-of-the-art explainability mechanics are employed to highlight the significance of each feature, and common patterns and differences with the medical bibliography are then discussed. Moreover, the findings are compared with common risk factors for CAD, to offer an evaluation of the prediction process from the medical expert’s point of view. By depicting how the algorithm weights the information contained in features, we shed light on the black-box mechanics of ML prediction models; by analyzing the findings, we explore their validity in accordance with the medical literature on the matter.
41
Rutten, Daan, and Debankur Mukherjee. "Capacity Scaling Augmented With Unreliable Machine Learning Predictions." ACM SIGMETRICS Performance Evaluation Review 49, no. 2 (January 17, 2022): 24–26. http://dx.doi.org/10.1145/3512798.3512808.
Full textAbstract:
Modern data centers suffer from immense power consumption. As a result, data center operators have heavily invested in capacity scaling solutions, which dynamically deactivate servers if the demand is low and activate them again when the workload increases. We analyze a continuoustime model for capacity scaling, where the goal is to minimize the weighted sum of flow-time, switching cost, and power consumption in an online fashion. We propose a novel algorithm, called the Adaptive Balanced Capacity Scaling (ABCS) algorithm, that has access to black-box machine learning predictions. ABCS aims to adapt to the predictions and is also robust against unpredictable surges in the workload. In particular, we prove that the ABCS algorithm is (1 + ")-competitive if the predictions are accurate, and yet, it has a uniformly bounded competitive ratio even if the predictions are completely inaccurate.
42
Ott, Simon, Adriano Barbosa-Silva, and Matthias Samwald. "LinkExplorer: predicting, explaining and exploring links in large biomedical knowledge graphs." Bioinformatics 38, no. 8 (February 9, 2022): 2371–73. http://dx.doi.org/10.1093/bioinformatics/btac068.
Full textAbstract:
Abstract Summary Machine learning algorithms for link prediction can be valuable tools for hypothesis generation. However, many current algorithms are black boxes or lack good user interfaces that could facilitate insight into why predictions are made. We present LinkExplorer, a software suite for predicting, explaining and exploring links in large biomedical knowledge graphs. LinkExplorer integrates our novel, rule-based link prediction engine SAFRAN, which was recently shown to outcompete other explainable algorithms and established black-box algorithms. Here, we demonstrate highly competitive evaluation results of our algorithm on multiple large biomedical knowledge graphs, and release a web interface that allows for interactive and intuitive exploration of predicted links and their explanations. Availability and implementation A publicly hosted instance, source code and further documentation can be found at https://github.com/OpenBioLink/Explorer. Supplementary information Supplementary data are available at Bioinformatics online.
43
Wang, Yanan, Xuebing Han, Languang Lu, Yangquan Chen, and Minggao Ouyang. "Sensitivity of Fractional-Order Recurrent Neural Network with Encoded Physics-Informed Battery Knowledge." Fractal and Fractional 6, no. 11 (November 2, 2022): 640. http://dx.doi.org/10.3390/fractalfract6110640.
Full textAbstract:
In the field of state estimation for the lithium-ion battery (LIB), model-based methods (white box) have been developed to explain battery mechanism and data-driven methods (black box) have been designed to learn battery statistics. Both white box methods and black box methods have drawn much attention recently. As the combination of white box and black box, physics-informed machine learning has been investigated by embedding physic laws. For LIB state estimation, this work proposes a fractional-order recurrent neural network (FORNN) encoded with physics-informed battery knowledge. Three aspects of FORNN can be improved by learning certain physics-informed knowledge. Firstly, the fractional-order state feedback is achieved by introducing a fractional-order derivative in a forward propagation process. Secondly, the fractional-order constraint is constructed by a voltage partial derivative equation (PDE) deduced from the battery fractional-order model (FOM). Thirdly, both the fractional-order gradient descent (FOGD) and fractional-order gradient descent with momentum (FOGDm) methods are proposed by introducing a fractional-order gradient in the backpropagation process. For the proposed FORNN, the sensitivity of the added fractional-order parameters are analyzed by experiments under the federal urban driving schedule (FUDS) operation conditions. The experiment results demonstrate that a certain range of every fractional-order parameter can achieve better convergence speed and higher estimation accuracy. On the basis of the sensitivity analysis, the fractional-order parameter tuning rules have been concluded and listed in the discussion part to provide useful references to the parameter tuning of the proposed algorithm.
44
Kammüller, Florian, and Dimpy Satija. "Explanation of Student Attendance AI Prediction with the Isabelle Infrastructure Framework." Information 14, no. 8 (August 10, 2023): 453. http://dx.doi.org/10.3390/info14080453.
Full textAbstract:
Right from the beginning, attendance has played an important role in the education systems, not only in student success but in the overall interest of the matter. Although all schools try to accentuate good attendance, still some schools find it hard to achieve the required level (96% in UK) of average attendance. The most productive way of increasing the pupils′ attendance rate is to predict when it is going to go down, understand the reasons—why it happened—and act on the affecting factors so as to prevent it. Artificial intelligence (AI) is an automated machine learning solution for different types of problems. Several machine learning (ML) models like logistic regression, decision trees, etc. are easy to understand; however, complicated (Neural Network, BART etc.) ML models are not transparent but are black-boxes for humans. It is not always evident how machine intelligence arrived at a decision. However, not always, but in critical applications it is important that humans can understand the reasons for such decisions. In this paper, we present a methodology on the application example of pupil attendance for constructing explanations for AI classification algorithms. The methodology includes building a model of the application in the Isabelle Insider and Infrastructure framework (IIIf) and an algorithm (PCR) that helps us to obtain a detailed logical rule to specify the performance of the black-box algorithm, hence allowing us to explain it. The explanation is provided within the logical model of the IIIf, thus is suitable for human audiences. It has been shown that the RR-cycle of IIIf can be adapted to provide a method for iteratively extracting an explanation by interleaving attack tree analysis with precondition refinement, which finally yields a general rule that describes the decision taken by a black-box algorithm produced by Artificial intelligence.
45
Salih, Dhiadeen Mohammed, Samsul Bahari Mohd Noor, Mohammad Hamiruce Merhaban, and Raja Mohd Kamil. "Wavelet Network: Online Sequential Extreme Learning Machine for Nonlinear Dynamic Systems Identification." Advances in Artificial Intelligence 2015 (September 20, 2015): 1–10. http://dx.doi.org/10.1155/2015/184318.
Full textAbstract:
A single hidden layer feedforward neural network (SLFN) with online sequential extreme learning machine (OSELM) algorithm has been introduced and applied in many regression problems successfully. However, using SLFN with OSELM as black-box for nonlinear system identification may lead to building models for the identified plant with inconsistency responses from control perspective. The reason can refer to the random initialization procedure of the SLFN hidden node parameters with OSELM algorithm. In this paper, a single hidden layer feedforward wavelet network (WN) is introduced with OSELM for nonlinear system identification aimed at getting better generalization performances by reducing the effect of a random initialization procedure.
46
Luong, Ngoc Hoang, Han La Poutré, and Peter A. N. Bosman. "Exploiting Linkage Information and Problem-Specific Knowledge in Evolutionary Distribution Network Expansion Planning." Evolutionary Computation 26, no. 3 (September 2018): 471–505. http://dx.doi.org/10.1162/evco_a_00209.
Full textAbstract:
This article tackles the Distribution Network Expansion Planning (DNEP) problem that has to be solved by distribution network operators to decide which, where, and/or when enhancements to electricity networks should be introduced to satisfy the future power demands. Because of many real-world details involved, the structure of the problem is not exploited easily using mathematical programming techniques, for which reason we consider solving this problem with evolutionary algorithms (EAs). We compare three types of EAs for optimizing expansion plans: the classic genetic algorithm (GA), the estimation-of-distribution algorithm (EDA), and the Gene-pool Optimal Mixing Evolutionary Algorithm (GOMEA). Not fully knowing the structure of the problem, we study the effect of linkage learning through the use of three linkage models: univariate, marginal product, and linkage tree. We furthermore experiment with the impact of incorporating different levels of problem-specific knowledge in the variation operators. Experiments show that the use of problem-specific variation operators is far more important for the classic GA to find high-quality solutions. In all EAs, the marginal product model and its linkage learning procedure have difficulty in capturing and exploiting the DNEP problem structure. GOMEA, especially when combined with the linkage tree structure, is found to have the most robust performance by far, even when an out-of-the-box variant is used that does not exploit problem-specific knowledge. Based on experiments, we suggest that when selecting optimization algorithms for power system expansion planning problems, EAs that have the ability to effectively model and efficiently exploit problem structures, such as GOMEA, should be given priority, especially in the case of black-box or grey-box optimization.
47
Yiğit, Tuncay, Nilgün Şengöz, Özlem Özmen, Jude Hemanth, and Ali Hakan Işık. "Diagnosis of Paratuberculosis in Histopathological Images Based on Explainable Artificial Intelligence and Deep Learning." Traitement du Signal 39, no. 3 (June 30, 2022): 863–69. http://dx.doi.org/10.18280/ts.390311.
Full textAbstract:
Artificial intelligence holds great promise in medical imaging, especially histopathological imaging. However, artificial intelligence algorithms cannot fully explain the thought processes during decision-making. This situation has brought the problem of explainability, i.e., the black box problem, of artificial intelligence applications to the agenda: an algorithm simply responds without stating the reasons for the given images. To overcome the problem and improve the explainability, explainable artificial intelligence (XAI) has come to the fore, and piqued the interest of many researchers. Against this backdrop, this study examines a new and original dataset using the deep learning algorithm, and visualizes the output with gradient-weighted class activation mapping (Grad-CAM), one of the XAI applications. Afterwards, a detailed questionnaire survey was conducted with the pathologists on these images. Both the decision-making processes and the explanations were verified, and the accuracy of the output was tested. The research results greatly help pathologists in the diagnosis of paratuberculosis.
48
Du, Xiaohu, Jie Yu, Zibo Yi, Shasha Li, Jun Ma, Yusong Tan, and Qinbo Wu. "A Hybrid Adversarial Attack for Different Application Scenarios." Applied Sciences 10, no. 10 (May 21, 2020): 3559. http://dx.doi.org/10.3390/app10103559.
Full textAbstract:
Adversarial attack against natural language has been a hot topic in the field of artificial intelligence security in recent years. It is mainly to study the methods and implementation of generating adversarial examples. The purpose is to better deal with the vulnerability and security of deep learning systems. According to whether the attacker understands the deep learning model structure, the adversarial attack is divided into black-box attack and white-box attack. In this paper, we propose a hybrid adversarial attack for different application scenarios. Firstly, we propose a novel black-box attack method of generating adversarial examples to trick the word-level sentiment classifier, which is based on differential evolution (DE) algorithm to generate semantically and syntactically similar adversarial examples. Compared with existing genetic algorithm based adversarial attacks, our algorithm can achieve a higher attack success rate while maintaining a lower word replacement rate. At the 10% word substitution threshold, we have increased the attack success rate from 58.5% to 63%. Secondly, when we understand the model architecture and parameters, etc., we propose a white-box attack with gradient-based perturbation against the same sentiment classifier. In this attack, we use a Euclidean distance and cosine distance combined metric to find the most semantically and syntactically similar substitution, and we introduce the coefficient of variation (CV) factor to control the dispersion of the modified words in the adversarial examples. More dispersed modifications can increase human imperceptibility and text readability. Compared with the existing global attack, our attack can increase the attack success rate and make modification positions in generated examples more dispersed. We’ve increased the global search success rate from 75.8% to 85.8%. Finally, we can deal with different application scenarios by using these two attack methods, that is, whether we understand the internal structure and parameters of the model, we can all generate good adversarial examples.
49
Saleem, Sobia, Marcus Gallagher, and Ian Wood. "Direct Feature Evaluation in Black-Box Optimization Using Problem Transformations." Evolutionary Computation 27, no. 1 (March 2019): 75–98. http://dx.doi.org/10.1162/evco_a_00247.
Full textAbstract:
Exploratory Landscape Analysis provides sample-based methods to calculate features of black-box optimization problems in a quantitative and measurable way. Many problem features have been proposed in the literature in an attempt to provide insights into the structure of problem landscapes and to use in selecting an effective algorithm for a given optimization problem. While there has been some success, evaluating the utility of problem features in practice presents some significant challenges. Machine learning models have been employed as part of the evaluation process, but they may require additional information about the problems as well as having their own hyper-parameters, biases and experimental variability. As a result, extra layers of uncertainty and complexity are added into the experimental evaluation process, making it difficult to clearly assess the effect of the problem features. In this article, we propose a novel method for the evaluation of problem features which can be applied directly to individual or groups of features and does not require additional machine learning techniques or confounding experimental factors. The method is based on the feature's ability to detect a prior ranking of similarity in a set of problems. Analysis of Variance (ANOVA) significance tests are used to determine if the feature has successfully distinguished the successive problems in the set. Based on ANOVA test results, a percentage score is assigned to each feature for different landscape characteristics. Experimental results for twelve different features on four problem transformations demonstrate the method and provide quantitative evidence about the ability of different problem features to detect specific properties of problem landscapes.
50
Barkalov, Konstantin, Ilya Lebedev, and Evgeny Kozinov. "Acceleration of Global Optimization Algorithm by Detecting Local Extrema Based on Machine Learning." Entropy 23, no. 10 (September 28, 2021): 1272. http://dx.doi.org/10.3390/e23101272.
Full textAbstract:
This paper features the study of global optimization problems and numerical methods of their solution. Such problems are computationally expensive since the objective function can be multi-extremal, nondifferentiable, and, as a rule, given in the form of a “black box”. This study used a deterministic algorithm for finding the global extremum. This algorithm is based neither on the concept of multistart, nor nature-inspired algorithms. The article provides computational rules of the one-dimensional algorithm and the nested optimization scheme which could be applied for solving multidimensional problems. Please note that the solution complexity of global optimization problems essentially depends on the presence of multiple local extrema. In this paper, we apply machine learning methods to identify regions of attraction of local minima. The use of local optimization algorithms in the selected regions can significantly accelerate the convergence of global search as it could reduce the number of search trials in the vicinity of local minima. The results of computational experiments carried out on several hundred global optimization problems of different dimensionalities presented in the paper confirm the effect of accelerated convergence (in terms of the number of search trials required to solve a problem with a given accuracy).