Journal articles on the topic 'Contrastive Explanations'
To see the other types of publications on this topic, follow the link: Contrastive Explanations.
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 'Contrastive Explanations.'
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
Morton, Adam. "Mathematical Modelling and Contrastive Explanation." Canadian Journal of Philosophy Supplementary Volume 16 (1990): 251–70. http://dx.doi.org/10.1080/00455091.1990.10717228.
Full textAbstract:
This is an enquiry into flawed explanations. Most of the effort in studies of the concept of explanation, scientific or otherwise, has gone into the contrast between clear cases of explanation and clear non-explanations.
2
Lipton, Peter. "Contrastive Explanation." Royal Institute of Philosophy Supplement 27 (March 1990): 247–66. http://dx.doi.org/10.1017/s1358246100005130.
Full textAbstract:
According to a causal model of explanation, we explain phenomena by giving their causes or, where the phenomena are themselves causal regularities, we explain them by giving a mechanism linking cause and effect. If we explain why smoking causes cancer, we do not give the cause of this causal connection, but we do give the causal mechanism that makes it. The claim that to explain is to give a cause is not only natural and plausible, but it also avoids many of the objections to other accounts of explanation, such as the views that to explain is to give a reason to believe the phenomenon occurred, to somehow make the phenomenon familiar, or to give a Deductive-Nomological argument. Unlike the reason for belief account, a causal model makes a clear distinction between understanding why a phenomenon occurs and merely knowing that it does, and the model does so in a way that makes understanding unmysterious and objective. Understanding is not some sort of super-knowledge, but simply more knowledge: knowledge of the phenomenon and knowledge of its causal history. A causal model makes it clear how something can explain without itself being explained, and so avoids the regress of whys, since we can know a phenomenon's cause without knowing the cause of the cause. It also accounts for legitimate self-evidencing explanations, explanations where the phenomenon is an essential part of the evidence that the explanation is correct, so the explanation can not supply a non-circular reason for believing the phenomenon occurred. There is no barrier to knowing a cause through its effects and also knowing that it is their cause. The speed of recession of a star explains its observed red-shift, even though the shift is an essential part of the evidence for its speed of recession. The model also avoids the most serious objection to the familiarity view, which is that some phenomena are familiar yet not understood, since a phenomenon can be perfectly familiar, such as the blueness of the sky or the fact that the same side of the moon always faces the earth, even if we do not know its cause. Finally, a causal model avoids many of the objections to the Deductive-Nomological model. Ordinary explanations do not have to meet the requirements of the Deductive-Nomological model, because one does not need to give a law to give a cause, and one does not need to know a law to have good reason to believe that a cause is a cause. As for the notorious over-permissiveness of the Deductive-Nomological model, the reason recession explains red-shift but not conversely, is simply that causes explain effects but not conversely, and the reason a conjunction of laws does not explain its conjuncts is that conjunctions do not cause their conjuncts.
3
Khalifa, Kareem. "Contrastive Explanations as Social Accounts." Social Epistemology 24, no. 4 (October 2010): 263–84. http://dx.doi.org/10.1080/02691728.2010.506960.
Full text
4
Boulter, Stephen. "CONTRASTIVE EXPLANATIONS IN EVOLUTIONARY BIOLOGY." Ratio 25, no. 4 (November 6, 2012): 425–41. http://dx.doi.org/10.1111/j.1467-9329.2012.00555.x.
Full text
5
Glymour, Bruce. "Contrastive, Non-Probabilistic Statistical Explanations." Philosophy of Science 65, no. 3 (September 1998): 448–71. http://dx.doi.org/10.1086/392656.
Full text
6
Elzein, Nadine. "The demand for contrastive explanations." Philosophical Studies 176, no. 5 (February 22, 2018): 1325–39. http://dx.doi.org/10.1007/s11098-018-1065-z.
Full text
7
Levy, Neil. "Contrastive Explanations: A Dilemma for Libertarians." Dialectica 59, no. 1 (July 1, 2005): 51–61. http://dx.doi.org/10.1111/j.1746-8361.2005.01004.x.
Full text
8
Krarup, Benjamin, Senka Krivic, Daniele Magazzeni, Derek Long, Michael Cashmore, and David E. Smith. "Contrastive Explanations of Plans through Model Restrictions." Journal of Artificial Intelligence Research 72 (October 27, 2021): 533–612. http://dx.doi.org/10.1613/jair.1.12813.
Full textAbstract:
In automated planning, the need for explanations arises when there is a mismatch between a proposed plan and the user’s expectation. We frame Explainable AI Planning as an iterative plan exploration process, in which the user asks a succession of contrastive questions that lead to the generation and solution of hypothetical planning problems that are restrictions of the original problem. The object of the exploration is for the user to understand the constraints that govern the original plan and, ultimately, to arrive at a satisfactory plan. We present the results of a user study that demonstrates that when users ask questions about plans, those questions are usually contrastive, i.e. “why A rather than B?”. We use the data from this study to construct a taxonomy of user questions that often arise during plan exploration. Our approach to iterative plan exploration is a process of successive model restriction. Each contrastive user question imposes a set of constraints on the planning problem, leading to the construction of a new hypothetical planning problem as a restriction of the original. Solving this restricted problem results in a plan that can be compared with the original plan, admitting a contrastive explanation. We formally define model-based compilations in PDDL2.1 for each type of constraint derived from a contrastive user question in the taxonomy, and empirically evaluate the compilations in terms of computational complexity. The compilations were implemented as part of an explanation framework supporting iterative model restriction. We demonstrate its benefits in a second user study.
9
Fogelin, Lars. "Inference to the Best Explanation: A Common and Effective Form of Archaeological Reasoning." American Antiquity 72, no. 4 (October 2007): 603–26. http://dx.doi.org/10.2307/25470436.
Full textAbstract:
Processual and postprocessual archaeologists implicitly employ the same epistemological system to evaluate the worth of different explanations: inference to the best explanation. This is good since inference to the best explanation is the most effective epistemological approach to archaeological reasoning available. Underlying the logic of inference to the best explanation is the assumption that the explanation that accounts for the most evidence is also most likely to be true. This view of explanation often reflects the practice of archaeological reasoning better than either the hypothetico-deductive method or hermeneutics. This article explores the logic of inference to the best explanation and provides clear criteria to determine what makes one explanation better than another. Explanations that are empirically broad, general, modest, conservative, simple, testable, and address many perspectives are better than explanations that are not. This article also introduces a system of understanding explanation that emphasizes the role of contrastive pairings in the construction of specific explanations. This view of explanation allows for a better understanding of when, and when not, to engage in the testing of specific explanations.
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
Malandri, Lorenzo, Fabio Mercorio, Mario Mezzanzanica, Navid Nobani, and Andrea Seveso. "ContrXT: Generating contrastive explanations from any text classifier." Information Fusion 81 (May 2022): 103–15. http://dx.doi.org/10.1016/j.inffus.2021.11.016.
Full text
12
Venkatsubramaniam, Bhaskaran. "A Novel Approach to Explainable AI using Formal Concept Lattice." International Journal of Innovative Technology and Exploring Engineering 11, no. 7 (June 30, 2022): 36–48. http://dx.doi.org/10.35940/ijitee.g9992.0611722.
Full textAbstract:
Current approaches in explainable AI use an interpretable model to approximate a black box model or use gradient techniques to determine the salient parts of the input. While it is true that such approaches provide intuition about the black box model, the primary purpose of an explanation is to be exact at an individual instance and also from a global perspective, which is difficult to obtain using such model based approximations or from salient parts. On the other hand, traditional, deterministic approaches satisfy this primary purpose of explainability of being exact at an individual instance and globally, while posing a challenge to scale for large amounts of data. In this work, we propose a deterministic, novel approach to explainability using a formal concept lattice for classification problems, that reveal accurate explanations both globally and locally, including generation of similar and contrastive examples around an instance. This technique consists of preliminary lattice construction, synthetic data generation using implications from the preliminary lattice followed by actual lattice construction which is used to generate local, global, similar and contrastive explanations. Using sanity tests like Implementation Invariance, Input transformation Invariance, Model parameter randomization sensitivity and model-outcome relationship randomization sensitivity, its credibility is proven. Explanations from the lattice are compared to a white box model in order to prove its trustworthiness.
13
Karthik, Valmeekam, Sarath Sreedharan, Sailik Sengupta, and Subbarao Kambhampati. "RADAR-X: An Interactive Interface Pairing Contrastive Explanations with Revised Plan Suggestions." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 18 (May 18, 2021): 16051–53. http://dx.doi.org/10.1609/aaai.v35i18.18009.
Full textAbstract:
Automated Planning techniques can be leveraged to build effective decision support systems that assist the human-in-the-loop. Such systems must provide intuitive explanations when the suggestions made by these systems seem inexplicable to the human. In this regard, we consider scenarios where the user questions the system's suggestion by providing alternatives (referred to as foils). In response, we empower existing decision support technologies to engage in an interactive explanatory dialogue with the user and provide contrastive explanations based on user-specified foils to reach a consensus on proposed decisions. To provide contrastive explanations, we adapt existing techniques in Explainable AI Planning (XAIP). Furthermore, we use this dialog to elicit the user's latent preferences and propose three modes of interaction that use these preferences to provide revised plan suggestions. Finally, we showcase a decision support system that provides all these capabilities.
14
Niiniluoto, Ilkka. "Explanation by Idealized Theories." Kairos. Journal of Philosophy & Science 20, no. 1 (June 1, 2018): 43–63. http://dx.doi.org/10.2478/kjps-2018-0003.
Full textAbstract:
AbstractThe use of idealized scientific theories in explanations of empirical facts and regularities is problematic in two ways: they don’t satisfy the condition that the explanans is true, and they may fail to entail the explanandum. An attempt to deal with the latter problem was proposed by Hempel and Popper with their notion of approximate explanation. A more systematic perspective on idealized explanations was developed with the method of idealization and concretization by the Poznan school (Nowak, Krajewski) in the 1970s. If idealizational laws are treated as counterfactual conditionals, they can be true or truthlike, and the concretizations of such laws may increase their degree of truthlikeness. By replacing Hempel’s truth requirement with the condition that an explanatory theory is truthlike one can distinguish several important types of approximate, corrective, and contrastive explanations by idealized theories. The conclusions have important consequences for the debates about scientific realism and anti-realism.
15
Aguilar-Palacios, Carlos, Sergio Munoz-Romero, and Jose Luis Rojo-Alvarez. "Cold-Start Promotional Sales Forecasting Through Gradient Boosted-Based Contrastive Explanations." IEEE Access 8 (2020): 137574–86. http://dx.doi.org/10.1109/access.2020.3012032.
Full text
16
Campbell, Neil. "Self-forming actions, contrastive explanations, and the structure of the will." Synthese 197, no. 3 (March 8, 2018): 1225–40. http://dx.doi.org/10.1007/s11229-018-1749-0.
Full text
17
Woodcock, Claire, Brent Mittelstadt, Dan Busbridge, and Grant Blank. "The Impact of Explanations on Layperson Trust in Artificial Intelligence–Driven Symptom Checker Apps: Experimental Study." Journal of Medical Internet Research 23, no. 11 (November 3, 2021): e29386. http://dx.doi.org/10.2196/29386.
Full textAbstract:
Background Artificial intelligence (AI)–driven symptom checkers are available to millions of users globally and are advocated as a tool to deliver health care more efficiently. To achieve the promoted benefits of a symptom checker, laypeople must trust and subsequently follow its instructions. In AI, explanations are seen as a tool to communicate the rationale behind black-box decisions to encourage trust and adoption. However, the effectiveness of the types of explanations used in AI-driven symptom checkers has not yet been studied. Explanations can follow many forms, including why-explanations and how-explanations. Social theories suggest that why-explanations are better at communicating knowledge and cultivating trust among laypeople. Objective The aim of this study is to ascertain whether explanations provided by a symptom checker affect explanatory trust among laypeople and whether this trust is impacted by their existing knowledge of disease. Methods A cross-sectional survey of 750 healthy participants was conducted. The participants were shown a video of a chatbot simulation that resulted in the diagnosis of either a migraine or temporal arteritis, chosen for their differing levels of epidemiological prevalence. These diagnoses were accompanied by one of four types of explanations. Each explanation type was selected either because of its current use in symptom checkers or because it was informed by theories of contrastive explanation. Exploratory factor analysis of participants’ responses followed by comparison-of-means tests were used to evaluate group differences in trust. Results Depending on the treatment group, two or three variables were generated, reflecting the prior knowledge and subsequent mental model that the participants held. When varying explanation type by disease, migraine was found to be nonsignificant (P=.65) and temporal arteritis, marginally significant (P=.09). Varying disease by explanation type resulted in statistical significance for input influence (P=.001), social proof (P=.049), and no explanation (P=.006), with counterfactual explanation (P=.053). The results suggest that trust in explanations is significantly affected by the disease being explained. When laypeople have existing knowledge of a disease, explanations have little impact on trust. Where the need for information is greater, different explanation types engender significantly different levels of trust. These results indicate that to be successful, symptom checkers need to tailor explanations to each user’s specific question and discount the diseases that they may also be aware of. Conclusions System builders developing explanations for symptom-checking apps should consider the recipient’s knowledge of a disease and tailor explanations to each user’s specific need. Effort should be placed on generating explanations that are personalized to each user of a symptom checker to fully discount the diseases that they may be aware of and to close their information gap.
18
Zhang, Pengchong, and Suzanne Graham. "Vocabulary learning through listening: Comparing L2 explanations, teacher codeswitching, contrastive focus-on-form and incidental learning." Language Teaching Research 24, no. 6 (February 15, 2019): 765–84. http://dx.doi.org/10.1177/1362168819829022.
Full textAbstract:
This study explored the teaching and learning of vocabulary through listening among 137 senior high-school learners of English as a foreign language (EFL) in China. It compared different types of Lexical Focus-on-Form delivered to four treatment groups: post-listening vocabulary explanations in the L2; codeswitched explanations; explanations providing additional crosslinguistic information (Contrastive Focus-on-Form; CFoF); and no explanations (NE). It also investigated the impact of the intervention on learners’ listening comprehension. Learners completed aural vocabulary tests at pre-, post- and delayed post-test and listening assessments at pre- and post-test. For short- and long-term vocabulary acquisition, the three groups receiving explanations significantly outperformed the NE group. Gains for the CFoF group were significantly greater than for the L2 and Codeswitching groups, for both short-term and long-term learning. For listening comprehension, only the NE group made significant improvement from the pre-test to the post-test, as well as making significantly greater pre- to post-test improvement than the CFoF and the L2 groups did. The article concludes by discussing these findings in relation to theories of vocabulary acquisition and listening comprehension, as well as their pedagogical implications.
19
Sreedharan, Sarath, Siddharth Srivastava, and Subbarao Kambhampati. "Using state abstractions to compute personalized contrastive explanations for AI agent behavior." Artificial Intelligence 301 (December 2021): 103570. http://dx.doi.org/10.1016/j.artint.2021.103570.
Full text
20
Sokol, Kacper, and Peter Flach. "Desiderata for Interpretability: Explaining Decision Tree Predictions with Counterfactuals." Proceedings of the AAAI Conference on Artificial Intelligence 33 (July 17, 2019): 10035–36. http://dx.doi.org/10.1609/aaai.v33i01.330110035.
Full textAbstract:
Explanations in machine learning come in many forms, but a consensus regarding their desired properties is still emerging. In our work we collect and organise these explainability desiderata and discuss how they can be used to systematically evaluate properties and quality of an explainable system using the case of class-contrastive counterfactual statements. This leads us to propose a novel method for explaining predictions of a decision tree with counterfactuals. We show that our model-specific approach exploits all the theoretical advantages of counterfactual explanations, hence improves decision tree interpretability by decoupling the quality of the interpretation from the depth and width of the tree.
21
Apicella, A., F. Isgrò, R. Prevete, and G. Tamburrini. "Middle-Level Features for the Explanation of Classification Systems by Sparse Dictionary Methods." International Journal of Neural Systems 30, no. 08 (July 14, 2020): 2050040. http://dx.doi.org/10.1142/s0129065720500409.
Full textAbstract:
Machine learning (ML) systems are affected by a pervasive lack of transparency. The eXplainable Artificial Intelligence (XAI) research area addresses this problem and the related issue of explaining the behavior of ML systems in terms that are understandable to human beings. In many explanation of XAI approaches, the output of ML systems are explained in terms of low-level features of their inputs. However, these approaches leave a substantive explanatory burden with human users, insofar as the latter are required to map low-level properties into more salient and readily understandable parts of the input. To alleviate this cognitive burden, an alternative model-agnostic framework is proposed here. This framework is instantiated to address explanation problems in the context of ML image classification systems, without relying on pixel relevance maps and other low-level features of the input. More specifically, one obtains sets of middle-level properties of classification inputs that are perceptually salient by applying sparse dictionary learning techniques. These middle-level properties are used as building blocks for explanations of image classifications. The achieved explanations are parsimonious, for their reliance on a limited set of middle-level image properties. And they can be contrastive, because the set of middle-level image properties can be used to explain why the system advanced the proposed classification over other antagonist classifications. In view of its model-agnostic character, the proposed framework is adaptable to a variety of other ML systems and explanation problems.
22
Valmeekam, Karthik, Sarath Sreedharan, Sailik Sengupta, and Subbarao Kambhampati. "RADAR-X: An Interactive Mixed Initiative Planning Interface Pairing Contrastive Explanations and Revised Plan Suggestions." Proceedings of the International Conference on Automated Planning and Scheduling 32 (June 13, 2022): 508–17. http://dx.doi.org/10.1609/icaps.v32i1.19837.
Full textAbstract:
Decision support systems seek to enable informed decision-making. In the recent years, automated planning techniques have been leveraged to empower such systems to better aid the human-in-the-loop. The central idea for such decision support systems is to augment the capabilities of the human-in-the-loop with automated planning techniques and enhance the quality of decision-making. In addition to providing planning support, effective decision support systems must be able to provide intuitive explanations based on specific user queries for proposed decisions to its end users. Using this as motivation, we present our decision support system RADAR-X that showcases the ability to engage the user in an interactive explanatory dialogue by first enabling them to specify an alternative to a proposed decision (which we refer to as foils), and then providing contrastive explanations to these user-specified foils which helps the user understand why a specific plan was chosen over the alternative (or foil). Furthermore, the system uses this dialogue to elicit the user's latent preferences and provides revised plan suggestions through three different interaction strategies.
23
Darwiche, Adnan, and Chunxi Ji. "On the Computation of Necessary and Sufficient Explanations." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 5 (June 28, 2022): 5582–91. http://dx.doi.org/10.1609/aaai.v36i5.20498.
Full textAbstract:
The complete reason behind a decision is a Boolean formula that characterizes why the decision was made. This recently introduced notion has a number of applications, which include generating explanations, detecting decision bias and evaluating counterfactual queries. Prime implicants of the complete reason are known as sufficient reasons for the decision and they correspond to what is known as PI explanations and abductive explanations. In this paper, we refer to the prime implicates of a complete reason as necessary reasons for the decision. We justify this terminology semantically and show that necessary reasons correspond to what is known as contrastive explanations. We also study the computation of complete reasons for multi-class decision trees and graphs with nominal and numeric features for which we derive efficient, closed-form complete reasons. We further investigate the computation of shortest necessary and sufficient reasons for a broad class of complete reasons, which include the derived closed forms and the complete reasons for Sentential Decision Diagrams (SDDs). We provide an algorithm which can enumerate their shortest necessary reasons in output polynomial time. Enumerating shortest sufficient reasons for this class of complete reasons is hard even for a single reason. For this problem, we provide an algorithm that appears to be quite efficient as we show empirically.
24
Wang, Paul Y., Sainyam Galhotra, Romila Pradhan, and Babak Salimi. "Demonstration of generating explanations for black-box algorithms using Lewis." Proceedings of the VLDB Endowment 14, no. 12 (July 2021): 2787–90. http://dx.doi.org/10.14778/3476311.3476345.
Full textAbstract:
Explainable artificial intelligence (XAI) aims to reduce the opacity of AI-based decision-making systems, allowing humans to scrutinize and trust them. Unlike prior work that attributes the responsibility for an algorithm's decisions to its inputs as a purely associational concept, we propose a principled causality-based approach for explaining black-box decision-making systems. We present the demonstration of Lewis, a system that generates explanations for black-box algorithms at the global, contextual, and local levels, and provides actionable recourse for individuals negatively affected by an algorithm's decision. Lewis makes no assumptions about the internals of the algorithm except for the availability of its input-output data. The explanations generated by Lewis are based on probabilistic contrastive counterfactuals, a concept that can be traced back to philosophical, cognitive, and social foundations of theories on how humans generate and select explanations. We describe the system layout of Lewis wherein an end-user specifies the underlying causal model and Lewis generates explanations for particular use-cases, compares them with explanations generated by state-of-the-art approaches in XAI, and provides actionable recourse when applicable. Lewis has been developed as open-source software; the code and the demonstration video are available at lewis-system.github.io.
25
Balabanis, George, Anastasia Stathopoulou, and Jiayu Qiao. "Favoritism Toward Foreign and Domestic Brands: A Comparison of Different Theoretical Explanations." Journal of International Marketing 27, no. 2 (April 24, 2019): 38–55. http://dx.doi.org/10.1177/1069031x19837945.
Full textAbstract:
Five theoretical approaches can predict favoritism toward domestic and foreign brands. This article applies a contrastive perspective to examine social identity, personal identity, cultural identity, system justification, and categorical cognition theories and their attendant constructs. The authors propose a set of main-effects hypotheses as well as hypotheses related to both product and country moderation effects on attitudes toward and loyalty to domestic and foreign brands. They test the hypotheses on a sample of Chinese consumers with respect to salient brands from 12 product categories. The results indicate that three of the theoretical approaches examined can explain only one side of favoritism—most commonly favoritism toward domestic brands—but not favoritism toward both domestic and foreign brands. Consumer xenocentrism, a concept rooted in system justification theory, seems to provide more consistent predictions for both domestic- and foreign-brand bias.
26
Jacovi, Alon, and Yoav Goldberg. "Aligning Faithful Interpretations with their Social Attribution." Transactions of the Association for Computational Linguistics 9 (2021): 294–310. http://dx.doi.org/10.1162/tacl_a_00367.
Full textAbstract:
Abstract We find that the requirement of model interpretations to be faithful is vague and incomplete. With interpretation by textual highlights as a case study, we present several failure cases. Borrowing concepts from social science, we identify that the problem is a misalignment between the causal chain of decisions (causal attribution) and the attribution of human behavior to the interpretation (social attribution). We reformulate faithfulness as an accurate attribution of causality to the model, and introduce the concept of aligned faithfulness: faithful causal chains that are aligned with their expected social behavior. The two steps of causal attribution and social attribution together complete the process of explaining behavior. With this formalization, we characterize various failures of misaligned faithful highlight interpretations, and propose an alternative causal chain to remedy the issues. Finally, we implement highlight explanations of the proposed causal format using contrastive explanations.
27
Sokol, Kacper, and Peter Flach. "One Explanation Does Not Fit All." KI - Künstliche Intelligenz 34, no. 2 (February 4, 2020): 235–50. http://dx.doi.org/10.1007/s13218-020-00637-y.
Full textAbstract:
Abstract The need for transparency of predictive systems based on Machine Learning algorithms arises as a consequence of their ever-increasing proliferation in the industry. Whenever black-box algorithmic predictions influence human affairs, the inner workings of these algorithms should be scrutinised and their decisions explained to the relevant stakeholders, including the system engineers, the system’s operators and the individuals whose case is being decided. While a variety of interpretability and explainability methods is available, none of them is a panacea that can satisfy all diverse expectations and competing objectives that might be required by the parties involved. We address this challenge in this paper by discussing the promises of Interactive Machine Learning for improved transparency of black-box systems using the example of contrastive explanations—a state-of-the-art approach to Interpretable Machine Learning. Specifically, we show how to personalise counterfactual explanations by interactively adjusting their conditional statements and extract additional explanations by asking follow-up “What if?” questions. Our experience in building, deploying and presenting this type of system allowed us to list desired properties as well as potential limitations, which can be used to guide the development of interactive explainers. While customising the medium of interaction, i.e., the user interface comprising of various communication channels, may give an impression of personalisation, we argue that adjusting the explanation itself and its content is more important. To this end, properties such as breadth, scope, context, purpose and target of the explanation have to be considered, in addition to explicitly informing the explainee about its limitations and caveats. Furthermore, we discuss the challenges of mirroring the explainee’s mental model, which is the main building block of intelligible human–machine interactions. We also deliberate on the risks of allowing the explainee to freely manipulate the explanations and thereby extracting information about the underlying predictive model, which might be leveraged by malicious actors to steal or game the model. Finally, building an end-to-end interactive explainability system is a challenging engineering task; unless the main goal is its deployment, we recommend “Wizard of Oz” studies as a proxy for testing and evaluating standalone interactive explainability algorithms.
28
Kenny, Eoin M., and Mark T. Keane. "On Generating Plausible Counterfactual and Semi-Factual Explanations for Deep Learning." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 13 (May 18, 2021): 11575–85. http://dx.doi.org/10.1609/aaai.v35i13.17377.
Full textAbstract:
There is a growing concern that the recent progress made in AI, especially regarding the predictive competence of deep learning models, will be undermined by a failure to properly explain their operation and outputs. In response to this disquiet, counterfactual explanations have become very popular in eXplainable AI (XAI) due to their asserted computational, psychological, and legal benefits. In contrast however, semi-factuals (which appear to be equally useful) have surprisingly received no attention. Most counterfactual methods address tabular rather than image data, partly because the non-discrete nature of images makes good counterfactuals difficult to define; indeed, generating plausible counterfactual images which lie on the data manifold is also problematic. This paper advances a novel method for generating plausible counterfactuals and semi-factuals for black-box CNN classifiers doing computer vision. The present method, called PlausIble Exceptionality-based Contrastive Explanations (PIECE), modifies all “exceptional” features in a test image to be “normal” from the perspective of the counterfactual class, to generate plausible counterfactual images. Two controlled experiments compare this method to others in the literature, showing that PIECE generates highly plausible counterfactuals (and the best semi-factuals) on several benchmark measures.
29
Marquez, Marc Kenneth Labadan. "A Contrastive Rhetoric Analysis of Professional Register in Internal Communication Electronic Mails." International Journal of Linguistics 8, no. 2 (April 21, 2016): 106. http://dx.doi.org/10.5296/ijl.v8i2.9348.
Full textAbstract:
<p>This study employs a corpus-based approach to identify and examine professional register features and some cultural-rhetorical patterns from a corpus of 50 internal communication electronic mails (e-mails) randomly culled from one Filipino and two American companies. Using the ten linguistic parameters in register variation in professional communication, similarities and differences in professional register features, as well as fluctuation tendencies, have been accounted. The findings have revealed that both e-mail corpora from the two language communities contained features significantly marked by professional casual register. However, a close inspection of the individual parameter frequency results has revealed considerable differences including register fluctuation tendencies, conformities to genre norms and conventionalities, and some culture-related rhetorical peculiarities. Moreover, the study has provided explanations on the importance of understanding rhetorical differences across cultures, as well as suggestions for further research endeavors on the given genre and language research field.</p>
30
Cronin, Sophie L., Morgan L. Spence, Paul A. Miller, and Derek H. Arnold. "Bidirectional Gender Face Aftereffects: Evidence Against Normative Facial Coding." Perception 46, no. 2 (October 22, 2016): 119–38. http://dx.doi.org/10.1177/0301006616672578.
Full textAbstract:
Facial appearance can be altered, not just by restyling but also by sensory processes. Exposure to a female face can, for instance, make subsequent faces look more masculine than they would otherwise. Two explanations exist. According to one, exposure to a female face renormalizes face perception, making that female and all other faces look more masculine as a consequence—a unidirectional effect. According to that explanation, exposure to a male face would have the opposite unidirectional effect. Another suggestion is that face gender is subject to contrastive aftereffects. These should make some faces look more masculine than the adaptor and other faces more feminine—a bidirectional effect. Here, we show that face gender aftereffects are bidirectional, as predicted by the latter hypothesis. Images of real faces rated as more and less masculine than adaptors at baseline tended to look even more and less masculine than adaptors post adaptation. This suggests that, rather than mental representations of all faces being recalibrated to better reflect the prevailing statistics of the environment, mental operations exaggerate differences between successive faces, and this can impact facial gender perception.
31
Thrun, Michael C., Alfred Ultsch, and Lutz Breuer. "Explainable AI Framework for Multivariate Hydrochemical Time Series." Machine Learning and Knowledge Extraction 3, no. 1 (February 4, 2021): 170–204. http://dx.doi.org/10.3390/make3010009.
Full textAbstract:
The understanding of water quality and its underlying processes is important for the protection of aquatic environments. With the rare opportunity of access to a domain expert, an explainable AI (XAI) framework is proposed that is applicable to multivariate time series. The XAI provides explanations that are interpretable by domain experts. In three steps, it combines a data-driven choice of a distance measure with supervised decision trees guided by projection-based clustering. The multivariate time series consists of water quality measurements, including nitrate, electrical conductivity, and twelve other environmental parameters. The relationships between water quality and the environmental parameters are investigated by identifying similar days within a cluster and dissimilar days between clusters. The framework, called DDS-XAI, does not depend on prior knowledge about data structure, and its explanations are tendentially contrastive. The relationships in the data can be visualized by a topographic map representing high-dimensional structures. Two state of the art XAIs called eUD3.5 and iterative mistake minimization (IMM) were unable to provide meaningful and relevant explanations from the three multivariate time series data. The DDS-XAI framework can be swiftly applied to new data. Open-source code in R for all steps of the XAI framework is provided and the steps are structured application-oriented.
32
Shahabi, Mitra, and Maria Teresa Roberto. "Metaphorical application and interpretation of animal terms." Languages in Contrast 15, no. 2 (November 6, 2015): 280–93. http://dx.doi.org/10.1075/lic.15.2.06sha.
Full textAbstract:
The present research adopts a contrastive and descriptive approach aiming at discovering the reason for similarities and differences between the metaphorical meanings of animal terms between the two languages of English and Persian. For this purpose the most popular animal metaphors in both languages are compared and contrasted. The animals are mostly those with which we have close contact in our daily lives. It is believed that if we could learn how metaphors have originated across languages we could find some explanations for similarities and differences of the metaphorical meanings across languages and cultures. Contrasting the origins of metaphorical concepts is believed to be an appropriate framework for this goal. The results of this study reflect how English and Persian people conceptualize their surrounding world across cultures and how they lexicalize them. It is found that although the physical characteristics and behaviour of animals are the basis for the metaphorical applications or interpretations of animal terms, they are not the only determining factor. The other factors in metaphorical meanings of animal names are culture, language-specificity, and also those behavioural characteristics of animals which are attributed to culture (culturally salient features).
33
WŁODARCZYK, JAROSŁAW. "THE PROBLEM OF LEXICAL GAPS IN TEACHING MILITARY ENGLISH." CONTEMPORARY MILITARY CHALLENGES 2022, no. 24/3 (September 30, 2022): 83–100. http://dx.doi.org/10.33179/bsv.99.svi.11.cmc.24.4.5.
Full textAbstract:
Since anisomorphism, the absence of an exact correspondence between words in two different languages, often impedes the process of second language (L2) acquisition, L2 teachers need to recognize effective methods for dealing with it. In an experiment involving a total of 109 students, we tested several methods of teaching English language military vocabulary particulars which lack direct Polish language equivalents. The results suggest that L1 translation is less effective in dealing with lexical gaps than monolingual explanations, presentation of the terminology in context, and illustrating its meaning with examples. However, the results also indicate that the use of L1 for contrastive analysis may help students cope with anisomorphism. Key words English for specific purposes, military terminology, vocabulary teaching, pomenska neprekrivnost, anisomorphism.
34
Wales, M. L. "The relative frequency of the synthetic and composite futures in the newspaper Ouest-France and some observations on distribution." Journal of French Language Studies 12, no. 1 (March 2002): 73–93. http://dx.doi.org/10.1017/s0959269502000157.
Full textAbstract:
Figures on the relative frequency of synthetic and composite future forms in Ouest-France are presented and compared with those of earlier studies on the passé simple and passé composé. The synthetic future is found to be dominant. Possible formal explanations for distribution are found to be inconclusive. Distribution across different text-types is found to be more promising, since contrastive functions of the two forms can be identified in texts where they co-occur. The composite future typically reports new proposals or plans as current news, while the synthetic future outlines details that will be realised at the time of implementation. Both functions are important in dailies, but current news is more often expressed in the present tense at the expense of the composite future.
35
Abenova, P. "LINGUOCULTURALASPECTSOFTRANSLATION OF MODERNAMERICANGONZO-REPORTING." BULLETIN Series of Philological Sciences 72, no. 2 (June 30, 2020): 641–47. http://dx.doi.org/10.51889/2020-2.1728-7804.103.
Full textAbstract:
This article discusses linguistic and cultural aspects of translation of modern American gonzo-reporting on the example of the book written by Thompson Hunter, who is father founder of genre “gonzo”. The genre “gonzo” appeared first in the USA within the new branch of journalism, which is called “New Journalism” and it developed there as well. Gonzo-reporting is a type of a text, where author uses subjective 1st person narration. He actively participates in the event that he describes. Gonzo-text has its specific stylistic and linguistic peculiarities, due to the fact that author speaks mind freely, uses humor, sarcasm, irony, jargons, slangs etc. Gonzo-texts contain deep descriptions of many details, explanations to author’s actions in reality. The translator’s main aim in translation of gonzo-reporting is to conduct adequate translation form linguistic and cultural aspects.The article provides a comparative and contrastive analysis of translations and the original; it discusses various approaches and criticism to the translation text.
36
Tabak, Iris, and Brian J. Reiser. "Software-realized inquiry support for cultivating a disciplinary stance." Pragmatics and Cognition 16, no. 2 (July 24, 2008): 307–55. http://dx.doi.org/10.1075/pc.16.2.06tab.
Full textAbstract:
What role can technology play in cultivating a disciplinary stance — raising questions, planning investigations, interpreting data and constructing explanations in a way that reflects disciplinary values and principles? How can overt and tacit expert scientific knowledge be captured, represented and used to design software that enables novices to assume a disciplinary stance in their investigations? We present The Galapagos Finches software designed to foster a biological and evolutionary stance. Our approach, Discipline-Specific Strategic Support (DSSS), translates the main variable types, comparison types and relationships in a discipline into manipulable objects in the interface. Pre/post-tests show how DSSS helps achieve a balance between content and process goals. A contrastive-case microanalysis of high, medium and low-achieving students’ inquiry shows progress toward a disciplinary stance. Our study shows how software representations carry multiple levels of meaning, and that the efficacy of learning technologies hinges on reflection at both the navigation and disciplinary-signification levels.
37
Tse, Holman. "Variation and change in Toronto heritage Cantonese." Asia-Pacific Language Variation 2, no. 2 (December 31, 2016): 124–56. http://dx.doi.org/10.1075/aplv.2.2.02tse.
Full textAbstract:
Abstract This paper presents the first sociophonetic study of Cantonese vowels using sociolinguistic interview data from the Heritage Language Variation and Change in Toronto Corpus. It focuses on four allophones [iː], [ɪk/ɪŋ], [uː], and [ʊk/ʊŋ] of two contrastive vowels /iː/ and /uː/ across two generations of speakers. The F1 and F2 of 30 vowel tokens were analyzed for these four allophones from each of 20 speakers (N = 600 vowel tokens). Results show inter-generational maintenance of allophonic conditioning for /iː/ and /uː/ as well as an interaction between generation and sex such that second-generation female speakers have the most retracted variants of [ɪk/ɪŋ] and the most fronted variants of [iː]. This paper will discuss three possible explanations based on internal motivation, phonetic assimilation, and phonological influence. This will illustrate the importance of multiple comparisons (including inter-generational, cross-linguistic, and cross-community) in the relatively new field of heritage language phonology research.
38
Blas Arroyo, José Luis. "When Language Contact Says Nothing: A Contrastive Analysis of Queísta Structures in Two Varieties of Peninsular Spanish." Journal of Language Contact 14, no. 2 (December 14, 2021): 403–37. http://dx.doi.org/10.1163/19552629-14020006.
Full textAbstract:
Abstract Based on the existence of some structural conflict between Spanish and Catalan in certain points of the syntax, this study tests the hypothesis about the influence of the latter on the distribution of queísmo uses (‘Me alegro que vengas’ [‘I’m glad you come’]) in the Spanish spoken in an eastern peninsular variety in contact with Catalan. Using the tools of comparative sociolinguistics, and the analysis of three corpora of contemporary Spanish, the study exhaustively examines the conditioning of this variable. The starting hypothesis is that the influence of the contact can be inferred from the comparison between different magnitudes derived from a multivariable statistical analysis. In addition to several linguistic and extra-linguistic predictors previously analysed in the literature, we also take into account other factor groups that may be particularly informative about that potential influence. Thus, from a structural point of view, we consider the contrast between: a) conjunctive queísmo in verbal structures, in which the structural conflict with Spanish is more evident (‘me acuerdo (de) que vino con su mujer/em recorde Ø que va vindre amb la seua dona’ [‘I remember that he came with his wife’]; and b) pronominal queísmo in relative sentences, in which the coincidence between both languages is greater (‘el día (en) que nos conocimos / el día (en) què ens vam conéixer’). From an extralinguistic perspective, the incidence of two additional factors is also examined: a) the speech community (without contact (Madrid/Alcalá) vs. in contact (Castellón), and b) the main language of the speakers (Spanish/Catalan-Valencian). The results of several mixed-effect regression analyses performed do not support the hypothesis of contact. The distributional differences between the above-mentioned groups are minimal, and in no case significant. On the other hand, the variation is basically affected by the same structural and non-structural predictors, regardless of the speech community or the ethnolinguistic group examined. Even the few divergences that are observed point in a direction contrary to that expected by the contact hypothesis. The study concludes with some potential explanations about these results and the contrast with other cases of syntactic convergence with Catalan.
39
Wallat, Cynthia, and Carolyn Steele. "Limitations of How We Categorize People." education policy analysis archives 7 (July 8, 1999): 21. http://dx.doi.org/10.14507/epaa.v7n21.1999.
Full textAbstract:
Social policy researchers and policy rules and regulation writers have not taken advantage of advances in assessing ways in which social representations of ideas about people can convey alternative explanations of social life. During the past decade a growing number of scholars have considered how representational practices and the representations that are outcomes of such practices have value. Neglecting to consider representational practices has consequences including failure to mobilize and sustain alternative ideologies that reject narrow perspectives on families and communities. As evidenced by recent OMB rulings on census categories, the dominant sense of meaning of population—and hence family and community—is quite similar to the 17th century sense of people as objects of a particular category in a place from which samples can be taken for statistical measurement. However, the contrastive analysis presented in this paper points out how sustained attention to consequences of use of sets of information categories collected to enumerate population to inform social policy can still materialize. In the wake of federal welfare reform, policy makers are particularly interested in questions of benefit relative to social service delivery and community revitalization. The presentation includes lessons learned from several dozen family, youth, school and community research projects.
40
Gijsbers, Victor. "Reconciling Contrastive and Non-contrastive Explanation." Erkenntnis 83, no. 6 (September 30, 2017): 1213–27. http://dx.doi.org/10.1007/s10670-017-9937-8.
Full text
41
McKenzie, Kevin. "Formulating professional identity." Pragmatics and Society 3, no. 1 (February 13, 2012): 31–60. http://dx.doi.org/10.1075/ps.3.1.02mck.
Full textAbstract:
Recent scholarly and practitioner research on the work of non-governmental organizations (NGOs) has been concerned with questions about the moral legitimacy of humanitarian aid in settings of armed conflict. At issue is the extent to which NGO activities are said to affect the conduct and outcome of warfare, thereby potentially implicating humanitarian aid in the partisan interests which it has traditionally eschewed as a condition of its legitimacy. This paper explores how such issues are taken up in the explanations offered by humanitarian aid operatives in descriptions of the work they carry out in settings of armed conflict. Drawing on a corpus of conversational material recorded in open-ended interviews with representatives of various NGOs that operate in Israel and the Occupied Palestinian Territories (OPT), I examine how speakers work to make themselves accountable to demands for sympathetic affiliation with the losing (or vanquished) parties in the Palestinian-Israeli conflict while maintaining a non-aligned stance relative to the partisan considerations that animate that conflict’s conduct. Both in first-hand narrative accounts of personal transformation and in descriptions of contrastive examples where professional colleagues are said to maintain a too-sympathetic affiliation with the partisan concerns of the Palestinian population whose needs they service, speakers work to provide for the legitimacy of their professional activities in the context of otherwise conflicting demands for moral accountability.
42
Came, Daniel. "Theism and Contrastive Explanation." European Journal for Philosophy of Religion 9, no. 1 (May 3, 2017): 19–26. http://dx.doi.org/10.24204/ejpr.v9i1.1862.
Full textAbstract:
I argue that there could not be grounds on which to introduce God into our ontology. My argument presupposes two doctrines. First, we should allow into our ontology only what figures in the best explanation of an event or fact. Second, explanation is contrastive by nature, in that the explanandum always consists in a contrast between a fact and a foil. I argue that God could not figure in true contrastive explanatory statements, because the omnipotence of God guarantees that for any true proposition p, God could have made it the case that ~p just as much as He could have made it the case that p.
43
Rieber, Steven. "Skepticism and Contrastive Explanation." Nous 32, no. 2 (June 1998): 189–204. http://dx.doi.org/10.1111/0029-4624.00096.
Full text
44
Cho, Young-mee Yu, Ji-Young Jung, and Minyoung Park. "Pedagogical Scenarios for Teaching Topic/Nominative Marking in Discourse." Korean Language in America 18, no. 1 (January 1, 2013): 40–70. http://dx.doi.org/10.2307/42922376.
Full textAbstract:
ABSTRACT Based on an error analysis of KFL learners' use of the topic marker (TM) and the nominative marker (NM), we will identify sources of common errors, define their semantically accurate and pragmatically appropriate uses within a pedagogical grammar of Korean, and then propose a series of pedagogical scenarios that facilitate acquisition of these markers. Most of the errors occurred in extended stretches of discourse rather than in isolated sentences, indicating that teaching discourse markers crucially needs pragmatic strategies that highlight target features in lieu of sentence-level, syntactic approaches. Linguistic analyses of the TM/NM in Korean have often treated them as analogous to their Japanese counterparts (Sohn, 1980); only recently have there been attempts to clarify important differences between the TM/NM in the two languages (E. Lee & Shimojo, 2011). Moreover, Korean pedagogical grammar treats the TM as: (1) a topic marker, (2) a contrastive marker, and (3) a marker of emphasis, and the NM as a subject marker, in much more simplistic terms Japanese pedagogical grammar (Noda, 1996). We propose that TM/NM (and bare nouns and zero pronominais) be taught in more concrete contexts with clear explanations of their discourse characteristics through activities designed to highlight their intended meanings and purposes. We believe that discourse marking should be introduced spirally along strategic points all the way up to advanced levels to reinforce its uses in complex structures and expand the learner's pragmatic repertoire of more subtle and interactional meanings.
45
Cho, Young-mee Yu, Ji-Young Jung, and Minyoung Park. "Pedagogical Scenarios for Teaching Topic/Nominative Marking in Discourse." Korean Language in America 18, no. 1 (January 1, 2013): 40–70. http://dx.doi.org/10.2307/korelangamer.18.2013.0040.
Full textAbstract:
ABSTRACT Based on an error analysis of KFL learners' use of the topic marker (TM) and the nominative marker (NM), we will identify sources of common errors, define their semantically accurate and pragmatically appropriate uses within a pedagogical grammar of Korean, and then propose a series of pedagogical scenarios that facilitate acquisition of these markers. Most of the errors occurred in extended stretches of discourse rather than in isolated sentences, indicating that teaching discourse markers crucially needs pragmatic strategies that highlight target features in lieu of sentence-level, syntactic approaches. Linguistic analyses of the TM/NM in Korean have often treated them as analogous to their Japanese counterparts (Sohn, 1980); only recently have there been attempts to clarify important differences between the TM/NM in the two languages (E. Lee & Shimojo, 2011). Moreover, Korean pedagogical grammar treats the TM as: (1) a topic marker, (2) a contrastive marker, and (3) a marker of emphasis, and the NM as a subject marker, in much more simplistic terms Japanese pedagogical grammar (Noda, 1996). We propose that TM/NM (and bare nouns and zero pronominais) be taught in more concrete contexts with clear explanations of their discourse characteristics through activities designed to highlight their intended meanings and purposes. We believe that discourse marking should be introduced spirally along strategic points all the way up to advanced levels to reinforce its uses in complex structures and expand the learner's pragmatic repertoire of more subtle and interactional meanings.
46
Pincock, Christopher. "Explanatory Relevance and Contrastive Explanation." Philosophy of Science 85, no. 5 (December 2018): 806–18. http://dx.doi.org/10.1086/699715.
Full text
47
Lipton, Peter. "Contrastive Explanation and Causal Triangulation." Philosophy of Science 58, no. 4 (December 1991): 687–97. http://dx.doi.org/10.1086/289648.
Full text
48
Witteveen, Joeri. "Natural Selection and Contrastive Explanation." Philosophy of Science 86, no. 3 (July 2019): 412–30. http://dx.doi.org/10.1086/703573.
Full text
49
Chien, Arnold. "Scalar implicature and contrastive explanation." Synthese 161, no. 1 (January 30, 2007): 47–66. http://dx.doi.org/10.1007/s11229-006-9153-6.
Full text
50
Khoutyz, Irina. "Storytelling in English and Russian-language lecture discourse." Moderna Språk 114, no. 2 (December 1, 2020): 73–90. http://dx.doi.org/10.58221/mosp.v114i2.7399.
Full textAbstract:
The advantages of using storytelling as a communication strategy are currently discussed in marketing, psychology, advertising, medicine, and education. It is described as an inborn skill that people relate to and which makes people listen. Storytelling is also described as an educational method (Dumović 2006, Pedersen 1995) and as a communication tool (Dahlstrom 2014, Sundin 2018) that can be used widely in education. The purpose of this paper is to analyze features and functions of storytelling used in academic lectures. For this purpose, three lecturers in Russian and three lecturers in English were analyzed. By means of a mixed-method approach, including discourse analysis and content analysis, it was established that all the lecturers used storytelling as a communication strategy for explanatory and contact-establishing purposes. The conclusion is made that storytelling was used by the Russian and the English-language lecturers to provide their audience with clear and engaging explanations. The lecturers constructed their stories using factual information, emotional and evaluative lexis, verbs of action, as well as discursive strategies that helped them to engage the audience in the topic of the lecture (e.g., the inclusive pronoun we and different kinds of questions). Though the stories have much in common, some differences between the stories in Russian and in English are identified. Based on these results, it is suggested that further contrastive studies of storytelling in academic settings can benefit those who are planning on presenting to an international audience or preparing to teach in an intercultural context.