Articoli di riviste sul tema "Counterfactual evaluation"
Segui questo link per vedere altri tipi di pubblicazioni sul tema: Counterfactual evaluation.
Cita una fonte nei formati APA, MLA, Chicago, Harvard e in molti altri stili
Vedi i top-50 articoli di riviste per l'attività di ricerca sul tema "Counterfactual evaluation".
Accanto a ogni fonte nell'elenco di riferimenti c'è un pulsante "Aggiungi alla bibliografia". Premilo e genereremo automaticamente la citazione bibliografica dell'opera scelta nello stile citazionale di cui hai bisogno: APA, MLA, Harvard, Chicago, Vancouver ecc.
Puoi anche scaricare il testo completo della pubblicazione scientifica nel formato .pdf e leggere online l'abstract (il sommario) dell'opera se è presente nei metadati.
Vedi gli articoli di riviste di molte aree scientifiche e compila una bibliografia corretta.
1
Markman, Keith D., Matthew N. McMullen, Ronald A. Elizaga e Nobuko Mizoguchi. "Counterfactual thinking and regulatory fit". Judgment and Decision Making 1, n. 2 (novembre 2006): 98–107. http://dx.doi.org/10.1017/s193029750000231x.
Abstract (sommario):
AbstractAccording to regulatory fit theory (Higgins, 2000), when people make decisions with strategies that sustain their regulatory focus orientation, they “feel right” about what they are doing, and this “feeling-right” experience then transfers to subsequent choices, decisions, and evaluations. The present research was designed to link the concept of regulatory fit to functional accounts of counterfactual thinking. In the present study, participants generated counterfactuals about their anagram performance, after which persistence on a second set of anagrams was measured. Under promotion framing (i.e., find 90% or more of all the possible words) upward counterfactual thinking in general elicited larger increases in persistence than did downward counterfactual thinking in general, but under prevention framing (i.e., avoid failing to find 90% or more of all the possible words) upward evaluation (comparing reality to a better reality) elicited larger increases in persistence than did upward reflection (focusing on a better reality), whereas downward reflection (focusing on a worse reality) elicited larger increases in persistence than did downward evaluation (comparing reality to a worse reality). In all, the present findings suggest that the generation of counterfactuals enhances the likelihood that individuals will engage in courses of action that fit with their regulatory focus orientation.
2
Bertolotti, Mauro, e Patrizia Catellani. "The Effects of Counterfactual Attacks on the Morality and Leadership of Different Professionals". Social Psychology 49, n. 3 (maggio 2018): 154–67. http://dx.doi.org/10.1027/1864-9335/a000338.
Abstract (sommario):
Abstract. Past research has offered contrasting results regarding the effects of attacks on social judgments. In three experiments, we investigated the effects of counterfactual (“If only…”) and non-counterfactual attacks on the morality versus leadership of politicians versus entrepreneurs. First, participants rated morality as the most desirable, but least typical dimension of politicians, and leadership as the most desirable and most typical dimension of entrepreneurs (Study 1). Then, counterfactual attacks led to poorer evaluation of the attacked target and better evaluation of the attacking source as compared to non-counterfactual attacks, especially when counterfactuals were focused on the most desirable dimension for the professional category of the attacked target (Study 2). Similar results emerged when the typicality of the attacked dimension was manipulated (Study 3). Discussion focuses on the higher success of attacks on desirable personality dimensions and of counterfactual attacks as compared to other attacks.
3
Hannikainen, Ivar. "Might-counterfactuals and the principle of conditional excluded middle". Disputatio 4, n. 30 (1 maggio 2011): 127–49. http://dx.doi.org/10.2478/disp-2011-0003.
Abstract (sommario):
Abstract Owing to the problem of inescapable clashes, epistemic accounts of might-counterfactuals have recently gained traction. In a different vein, the might argument against conditional excluded middle has rendered the latter a contentious principle to incorporate into a logic for conditionals. The aim of this paper is to rescue both ontic mightcounterfactuals and conditional excluded middle from these disparate debates and show them to be compatible. I argue that the antecedent of a might-counterfactual is semantically underdetermined with respect to the counterfactual worlds it selects for evaluation. This explains how might-counterfactuals select multiple counterfactual worlds as they apparently do and why their utterance confers a weaker alethic commitment on the speaker than does that of a would-counterfactual, as well as provides an ontic solution to inescapable clashes. I briefly sketch how the semantic underdetermination and truth conditions of mightcounterfactuals are regulated by conversational context.
4
Rylková, Žaneta, Karel Stelmach e Petr Vlček. "Overall Equipment Effectiveness within Counterfactual Impact Evaluation Concept". Scientific Annals of Economics and Business 64, s1 (1 dicembre 2017): 29–40. http://dx.doi.org/10.1515/saeb-2017-0037.
Abstract (sommario):
Abstract Counterfactual impact evaluation (CIE) is a scientific quantitative approach mainly based on experiments and quasi experiments. CIE is trying to prove a causal relationship between outputs and outcomes. CIE does not take into account coherence of external incentives of companies with internal incentives that have or may have an impact on the behaviour of enterprises. The paper sets up internal evaluation indicators for businesses, counterfactuals useful for creating a more complex metrics evaluating businesses in the area of performance. The aim of the paper is to present model situation using the elementary principle of counterfactual impact evaluation based on “the Overall Equipment Effectiveness (OEE)”.
5
Schleich, Maximilian, Zixuan Geng, Yihong Zhang e Dan Suciu. "GeCo". Proceedings of the VLDB Endowment 14, n. 9 (maggio 2021): 1681–93. http://dx.doi.org/10.14778/3461535.3461555.
Abstract (sommario):
Machine learning is increasingly applied in high-stakes decision making that directly affect people's lives, and this leads to an increased demand for systems to explain their decisions. Explanations often take the form of counterfactuals , which consists of conveying to the end user what she/he needs to change in order to improve the outcome. Computing counterfactual explanations is challenging, because of the inherent tension between a rich semantics of the domain, and the need for real time response. In this paper we present CeCo, the first system that can compute plausible and feasible counterfactual explanations in real time. At its core, CeCo relies on a genetic algorithm, which is customized to favor searching counterfactual explanations with the smallest number of changes. To achieve real-time performance, we introduce two novel optimizations: Δ-representation of candidate counterfactuals, and partial evaluation of the classifier. We compare empirically CeCo against five other systems described in the literature, and show that it is the only system that can achieve both high quality explanations and real time answers.
6
Wauters, Benedict, e Derek Beach. "Process tracing and congruence analysis to support theory-based impact evaluation". Evaluation 24, n. 3 (luglio 2018): 284–305. http://dx.doi.org/10.1177/1356389018786081.
Abstract (sommario):
Theory-based impact evaluations have been put forward increasingly as an alternative for counterfactual impact evaluations. However, this raises questions regarding the foundations of drawing causal inference on the basis of such approaches. Case study methods such as QCA (Quantitative Comparative Analysis), process tracing and congruence analysis are emerging as a way to match the methodological rigor of counterfactuals. While QCA relies on multiple cases, process tracing and congruence analysis are methods that claim to be able to draw causal inference within a single case. In this article, a completed theory-based impact evaluation of a European Social Fund intervention is used as a foundation to demonstrate and discuss the differences between process tracing and congruence analysis and their relative (dis)advantages.
7
Clark-Moorman, Kyleigh, Jason Rydberg e Edmund F. McGarrell. "Impact Evaluation of a Parolee-Based Focused Deterrence Program on Community-Level Violence". Criminal Justice Policy Review 30, n. 9 (27 novembre 2018): 1408–30. http://dx.doi.org/10.1177/0887403418812999.
Abstract (sommario):
We estimate the impact of a parolee-based focused deterrence (“pulling levers”) intervention on community-level firearm and non-firearm violence in Rockford, Illinois, via a retrospective, quasi-experimental design. Focusing on incidents of firearm violence in Rockford over a period of 60 months (38 months pre-intervention, 22 months post-intervention), program impact is assessed using Bayesian Structural Time Series (BSTS) models, constructing a synthetic control-based counterfactual time series from National Incident-Based Reporting System (NIBRS) data from 59 non-treated cities of similar size. Relative to the synthetic control counterfactual, the intervention was associated with significant reductions in both firearm and non-firearm violence, particularly robberies, ranging from 6% to 30%. Consistent with research at other sites, these findings support the notion that focused deterrence strategies centered on high-risk parolees may result in reductions in firearm violence at the community level. The BSTS approach is a useful application for producing counterfactuals in retrospective quasi-experimental impact evaluations.
8
Castaño, Javier, Maria Blanco e Pilar Martinez. "Reviewing Counterfactual Analyses to Assess Impacts of EU Rural Development Programmes: What Lessons Can Be Learned from the 2007–2013 Ex-Post Evaluations?" Sustainability 11, n. 4 (20 febbraio 2019): 1105. http://dx.doi.org/10.3390/su11041105.
Abstract (sommario):
Counterfactual analysis has been recommended as a means of assessing the impacts of European Rural Development Programmes (RDP) over recent years, although its application has been scarce to date. This paper examines the use of counterfactual analysis to assess socioeconomic impacts in a set of 2007–2013 ex-post evaluations. The analysis undertaken shows that a wide variety of counterfactual approaches have been applied, although certain barriers still remain to address the estimation of RDP impacts following the EU evaluation standards. Furthermore, we noted that impacts provided by individual RDP evaluations may hardly be aggregated, making it difficult to draw clear conclusions about the effectiveness of rural development policy at the EU level.
9
Zhang, Qiyuan, e Judith Covey. "Past and Future Implications of Near-Misses and Their Emotional Consequences". Experimental Psychology 61, n. 2 (1 ottobre 2014): 118–26. http://dx.doi.org/10.1027/1618-3169/a000231.
Abstract (sommario):
The Reflection and Evaluation Model (REM) of comparative thinking predicts that temporal perspective could moderate people’s emotional reactions to close counterfactuals following near-misses ( Markman & McMullen, 2003 ). The experiments reported in this paper tested predictions derived from this theory by examining how people’s emotional reactions to a near-miss at goal during a football match (Experiment 1) or a close score in a TV game show (Experiment 2) depended on the level of perceived future possibility. In support of the theory it was found that the presence of future possibility enhanced affective assimilation (e.g., if the near-miss occurred at the beginning of the game the players who had nearly scored were hopeful of future success) whereas the absence of future possibility enhanced affective contrast (e.g., if the near-miss occurred at the end of the game the players who had nearly scored were disappointed about missing an opportunity). Furthermore the experiments built upon our theoretical understanding by exploring the mechanisms which produce assimilation and contrast effects. In Experiment 1 we examined the incidence of present-oriented or future-oriented thinking, and in Experiment 2 we examined the mediating role of counterfactual thinking in the observed effect of proximity on emotions by testing whether stronger counterfactuals (measured using counterfactual probability estimates) produce bigger contrast and assimilation effects. While the results of these investigations generally support the REM, they also highlight the necessity to consider other psychological mechanisms (e.g., social comparison), in addition to counterfactual thinking, that might contribute to the emotional consequences of near-miss outcomes.
10
Snider, Todd, e Adam Bjorndahl. "Informative counterfactuals". Semantics and Linguistic Theory 25 (29 ottobre 2015): 1. http://dx.doi.org/10.3765/salt.v25i0.3077.
Abstract (sommario):
A single counterfactual conditional can have a multitude of interpretations that differ, intuitively, in the connection between antecedent and consequent. Using structural equation models (SEMs) to represent event dependencies, we illustrate various types of explanation compatible with a given counterfactual. We then formalize in the SEM framework the notion of an acceptable explanation, identifying the class of event dependencies compatible with a given counterfactual. Finally, by incorporating SEMs into possible worlds, we provide an update semantics with the enriched structure necessary for the evaluation of counterfactual conditionals.
11
Markman, Keith D., e Matthew N. McMullen. "Counterfactuals need not be comparative: The case of “As if”". Behavioral and Brain Sciences 30, n. 5-6 (dicembre 2007): 461–62. http://dx.doi.org/10.1017/s0140525x07002671.
Abstract (sommario):
AbstractByrne (2005) assumes that counterfactual thinking requires a comparison of facts with an imagined alternative. In our view, however, this assumption is unnecessarily restrictive. We argue that individuals do not necessarily engage in counterfactual simulations exclusively to evaluate factual reality. Instead, comparative evaluation is often suspended in favor of experiencing the counterfactual simulation as if it were real.
12
Ross, Stephen L., Eric Brunner e Rachel Rosen. "Identification and Counterfactuals for Program Evaluation of Career and Technical Education". Career and Technical Education Research 46, n. 3 (1 dicembre 2021): 15–36. http://dx.doi.org/10.5328/cter46.3.15.
Abstract (sommario):
This paper considers recent efforts to conduct experimental and quasi-experimental evaluations of career and technical education programs. It focuses on understanding the counterfactual, or control population, for these program evaluations, discussing how the educational experiences of the control population might vary from those of the treated population and the ways in which the treatment and control populations used for evaluation may differ from each other. The paper begins by discussing the key identification strategies and the associated assumptions used to identify program effects, including regression and propensity score matching, instrumental variables, regression discontinuity designs, randomized controlled trials, and lottery-based admissions. It then presents a series of case studies evaluating the counterfactual for specific studies that used each identification strategy.
13
Du, Zaichao, e Pei Pei. "A simple and robust counterfactual impact evaluation". Economics Letters 207 (ottobre 2021): 110015. http://dx.doi.org/10.1016/j.econlet.2021.110015.
14
Cummings, Rick. "‘What if’: The counterfactual in program evaluation". Evaluation Journal of Australasia 6, n. 2 (settembre 2006): 6–15. http://dx.doi.org/10.1177/1035719x0600600203.
15
Gabriel, Erin E., e Dean Follmann. "Augmented trial designs for evaluation of principal surrogates". Biostatistics 17, n. 3 (28 gennaio 2016): 453–67. http://dx.doi.org/10.1093/biostatistics/kxv055.
Abstract (sommario):
Abstract Observation of counterfactual intermediate responses, and evaluation of them as candidate surrogates, is complicated in a standard randomized trial as half of the responses are systematically missing by design. Although some augmentation procedures exist for obtaining counterfactual responses, they are specific to vaccine trials. We outline extensions to the existing augmentations and suggest augmentations of three trial designs outside the setting of vaccines. We outline the assumptions needed to identify the causal estimands of interest under each augmented design, under which standard likelihood-based methods can be used to evaluate intermediate responses as principal surrogates. Two of these designs, crossover and individual stepped-wedge, allow for the observation of clinical endpoints under both treatment and control for some subset of subjects and can therefore improve efficiency over standard parallel trial designs. The third, the treatment run-in design, allows for the observation of a baseline measure that may be as useful a surrogate as the true counterfactual intermediate response. As the evaluation methods rely on several assumptions, we also outline a remediation analysis, which can be used to help overcome assumption violations. We illustrate our suggested methods in an example from a drug-resistant tuberculosis treatment trial.
16
de Oliveira, Raphael Mazzine Barbosa, e David Martens. "A Framework and Benchmarking Study for Counterfactual Generating Methods on Tabular Data". Applied Sciences 11, n. 16 (7 agosto 2021): 7274. http://dx.doi.org/10.3390/app11167274.
Abstract (sommario):
Counterfactual explanations are viewed as an effective way to explain machine learning predictions. This interest is reflected by a relatively young literature with already dozens of algorithms aiming to generate such explanations. These algorithms are focused on finding how features can be modified to change the output classification. However, this rather general objective can be achieved in different ways, which brings about the need for a methodology to test and benchmark these algorithms. The contributions of this work are manifold: First, a large benchmarking study of 10 algorithmic approaches on 22 tabular datasets is performed, using nine relevant evaluation metrics; second, the introduction of a novel, first of its kind, framework to test counterfactual generation algorithms; third, a set of objective metrics to evaluate and compare counterfactual results; and, finally, insight from the benchmarking results that indicate which approaches obtain the best performance on what type of dataset. This benchmarking study and framework can help practitioners in determining which technique and building blocks most suit their context, and can help researchers in the design and evaluation of current and future counterfactual generation algorithms. Our findings show that, overall, there’s no single best algorithm to generate counterfactual explanations as the performance highly depends on properties related to the dataset, model, score, and factual point specificities.
17
Ferraro, Paul J. "Counterfactual thinking and impact evaluation in environmental policy". New Directions for Evaluation 2009, n. 122 (marzo 2009): 75–84. http://dx.doi.org/10.1002/ev.297.
18
Mellow, David. "Counterfactuals and the Proportionality Criterion". Ethics & International Affairs 20, n. 4 (dicembre 2006): 439–54. http://dx.doi.org/10.1111/j.1747-7093.2006.00044.x.
Abstract (sommario):
It is widely held that, in order for a resort to war or military force to be morally justified, it must, in addition to having a cause that is just, be proportionate. In this essay I argue for the need to use a counterfactual baseline when making the proportionality evaluation. Specifically, I argue that the relevant counterfactual baseline must contain a moral qualifier. In defending my proposal, I also contend that the relevant goods and harms that are weighed in the proportionality evaluation are not as open-ended as is sometimes presumed.
19
Lobachevsky, Yakov P., Alexander V. Shemyakin, Nikolay V. Limarenko, Ivan A. Uspensky e Ivan A. Yukhin. "Counterfactual Analysis of the Efficiency of Decontamination of Livestock Production Organic Wastes". Engineering Technologies and Systems 4, n. 33 (29 dicembre 2023): 466–89. http://dx.doi.org/10.15507/2658-4123.033.202304.466-489.
Abstract (sommario):
Introduction. The implementation of the decree of the President of the Russian Federation is aimed at ensuring the food security of the country and requires the industrialization of the agro-industrial sector. The effectiveness of industrialization depends on the use of automated, intelligent solutions at all stages of implementing technological processes. Livestock is an agro-industrial sector generating the largest amount of organic waste materials, which are potential energy carriers: litter, liquid manure, process effluents, etc. According to the data from the Russian Statistics Committee and the research results, the annual volume of manure generated from farms is from 43.3 to 45.1 million tons, while there is an upward trend. The used energy potential from the entire volume does not exceed 40%. It is possible to increase the efficiency of using the energy potential of organic animal waste materials through implementing digitalized solutions. A strategic tool for the effective industrialization of livestock is the implementation of application software products that ensure the growth of ecological and energy effects. Aim of the Article. The aim of the study is a counterfactual evaluation of the efficiency of the model for decontaminating liquid pig manure in the decontamination activator. Materials and Methods. Counterfactual analysis is a tool for formalizing complex, multifactorial processes to ensure their subsequent digitalization. The essence of the analysis consists in a “surveyˮ of the analyzed model through which the values of variables are determined providing changes that lead to a deviation of the response beyond the boundary conditions during interpretation. The advantage of counterfactual analysis is the stability and transparency of the model to external influences during machine learning. It is known that the representative pathogenic markers of the decontamination efficiency of liquid pig manure are helminth eggs and the number of colony-forming units of common coliform bacteria (CFU CCB). However, for testing and implementing an algorithm for counterfactual analysis of a mathematical model, it is acceptable to use the number of CFU CCB. The object of the study was liquid pig manure with a humidity from 88% to 98%, the subject was a counterfactual analysis of the dependence of the number of CFU CCB on the exposure time in the activator, the concentration of active chlorine, the mass of working bodies, magnetic induction, and liquid manure humidity. Results. The results of counterfactual evaluation and analysis carried with the use of the Python programming language and the PyCharm 2022.2 environment are presented in the tables. The counterfactual evaluation made it possible to identify ranges of variation of factors, the use of which can represent the potential of boundary conditions in solving the optimization problem. The cells of these values are highlighted in grey-blue. The most preferred ranges based on counterfactual evaluation are in the cells highlighted in green. Discussion and Conclusions. There has been substantiated the prospects of using active chlorine in combination with the influence of ferromagnetic working bodies moving in an alternating rotating electromagnetic field as a decontamination activator. On the basis of counterfactual evaluation it was established that the most significant factors for determining the efficiency of decontamination of liquid pig manure by the number of CFU CCB are: magnetic induction in the working zone of the activator inductor, active chlorine concentration and exposure time.
20
Chen, Jiangjie, Chun Gan, Sijie Cheng, Hao Zhou, Yanghua Xiao e Lei Li. "Unsupervised Editing for Counterfactual Stories". Proceedings of the AAAI Conference on Artificial Intelligence 36, n. 10 (28 giugno 2022): 10473–81. http://dx.doi.org/10.1609/aaai.v36i10.21290.
Abstract (sommario):
Creating what-if stories requires reasoning about prior statements and possible outcomes of the changed conditions. One can easily generate coherent endings under new conditions, but it would be challenging for current systems to do it with minimal changes to the original story. Therefore, one major challenge is the trade-off between generating a logical story and rewriting with minimal-edits. In this paper, we propose EDUCAT, an editing-based unsupervised approach for counterfactual story rewriting. EDUCAT includes a target position detection strategy based on estimating causal effects of the what-if conditions, which keeps the causal invariant parts of the story. EDUCAT then generates the stories under fluency, coherence and minimal-edits constraints. We also propose a new metric to alleviate the shortcomings of current automatic metrics and better evaluate the trade-off. We evaluate EDUCAT on a public counterfactual story rewriting benchmark. Experiments show that EDUCAT achieves the best trade-off over unsupervised SOTA methods according to both automatic and human evaluation. The resources of EDUCAT are available at: https://github.com/jiangjiechen/EDUCAT.
21
Rowe, Andy. "Rapid impact evaluation". Evaluation 25, n. 4 (19 settembre 2019): 496–513. http://dx.doi.org/10.1177/1356389019870213.
Abstract (sommario):
Rapid Impact Evaluation offers the potential to evaluate impacts in both ex ante and ex post settings, providing utility for developmental and formative evaluation as well as the usual summative settings. Rapid Impact Evaluation triangulates judgments of three separate groups of experts to assess the incremental change in effects attributable to the program. Three methodological innovations are central to the method: the scenario-based counterfactual, a simplified approach to measuring change in effects, and an interest-based approach to stakeholder engagement. In evaluations to date, Rapid Impact Evaluation has proved to be a cost effective and nimble approach to assessing impacts and does not intrude on design or implementation of the program. By applying recent thinking on use-seeking research emphasizing joint knowledge processes over knowledge products, Rapid Impact Evaluation promotes salience, legitimacy, and credibility with decision makers and key stakeholders. Applications show Rapid Impact Evaluation to be fit for purpose.
22
Huang, Wen, Lu Zhang e Xintao Wu. "Achieving Counterfactual Fairness for Causal Bandit". Proceedings of the AAAI Conference on Artificial Intelligence 36, n. 6 (28 giugno 2022): 6952–59. http://dx.doi.org/10.1609/aaai.v36i6.20653.
Abstract (sommario):
In online recommendation, customers arrive in a sequential and stochastic manner from an underlying distribution and the online decision model recommends a chosen item for each arriving individual based on some strategy. We study how to recommend an item at each step to maximize the expected reward while achieving user-side fairness for customers, i.e., customers who share similar profiles will receive a similar reward regardless of their sensitive attributes and items being recommended. By incorporating causal inference into bandits and adopting soft intervention to model the arm selection strategy, we first propose the d-separation based UCB algorithm (D-UCB) to explore the utilization of the d-separation set in reducing the amount of exploration needed to achieve low cumulative regret. Based on that, we then propose the fair causal bandit (F-UCB) for achieving the counterfactual individual fairness. Both theoretical analysis and empirical evaluation demonstrate effectiveness of our algorithms.
23
Yuan, Yuyu, Pengqian Zhao, Ting Guo e Hongpu Jiang. "Counterfactual-Based Action Evaluation Algorithm in Multi-Agent Reinforcement Learning". Applied Sciences 12, n. 7 (28 marzo 2022): 3439. http://dx.doi.org/10.3390/app12073439.
Abstract (sommario):
Multi-agent reinforcement learning (MARL) algorithms have made great achievements in various scenarios, but there are still many problems in solving sequential social dilemmas (SSDs). In SSDs, the agent’s actions not only change the instantaneous state of the environment but also affect the latent state which will, in turn, affect all agents. However, most of the current reinforcement learning algorithms focus on analyzing the value of instantaneous environment state while ignoring the study of the latent state, which is very important for establishing cooperation. Therefore, we propose a novel counterfactual reasoning-based multi-agent reinforcement learning algorithm to evaluate the continuous contribution of agent actions on the latent state. We compute that using simulation reasoning and building an action evaluation network. Then through counterfactual reasoning, we can get a single agent’s influence on the environment. Using this continuous contribution as an intrinsic reward enables the agent to consider the collective, thereby promoting cooperation. We conduct experiments in the SSDs environment, and the results show that the collective reward is increased by at least 25% which demonstrates the excellent performance of our proposed algorithm compared to the state-of-the-art algorithms.
24
Jacobone, Vittoria, Giuseppe Moro e Caterina Balenzano. "Counterfactual evaluation of a public programme for youth-led projects". RIV Rassegna Italiana di Valutazione, n. 75 (giugno 2021): 87–115. http://dx.doi.org/10.3280/riv2019-075006.
25
Pseiridis, Anastasia, e Ioannis Kostopoulos. "A Counterfactual Impact Evaluation of EU State Aid in Greece". WSEAS TRANSACTIONS ON BUSINESS AND ECONOMICS 20 (17 gennaio 2023): 352–72. http://dx.doi.org/10.37394/23207.2023.20.33.
Abstract (sommario):
EU state aid adopted from Member States is increasing at a fast pace due to the Covid-19 pandemic and energy crisis. Given its impact on the European economy, securing a maximum value added is a challenge for both policy makers and public administration. State aid impact depends not only on available resources but also on spending decisions that must be in line with state aid rules. It is believed that new policies would benefit if they were based on assessed evidence of existing policies during periods with similar characteristics. Our contribution analyses the characteristics of Greek development law based on a unique dataset extracted from the management information system of the Ministry of Economy. We hypothesize that there will be a change in firm productivity in the first years since program closure. Using counterfactual impact evaluation and propensity score matching, we find that there is a minor negative impact of development law on productivity. This might be an indication that firms receiving state aid do not perform as expected and perhaps better planning during policy modeling is needed.
26
Keogh, Ruth H., e Nan Van Geloven. "Prediction Under Interventions: Evaluation of Counterfactual Performance Using Longitudinal Observational Data". Epidemiology 35, n. 3 (18 aprile 2024): 329–39. http://dx.doi.org/10.1097/ede.0000000000001713.
Abstract (sommario):
Predictions under interventions are estimates of what a person’s risk of an outcome would be if they were to follow a particular treatment strategy, given their individual characteristics. Such predictions can give important input to medical decision-making. However, evaluating the predictive performance of interventional predictions is challenging. Standard ways of evaluating predictive performance do not apply when using observational data, because prediction under interventions involves obtaining predictions of the outcome under conditions that are different from those that are observed for a subset of individuals in the validation dataset. This work describes methods for evaluating counterfactual performance of predictions under interventions for time-to-event outcomes. This means we aim to assess how well predictions would match the validation data if all individuals had followed the treatment strategy under which predictions are made. We focus on counterfactual performance evaluation using longitudinal observational data, and under treatment strategies that involve sustaining a particular treatment regime over time. We introduce an estimation approach using artificial censoring and inverse probability weighting that involves creating a validation dataset mimicking the treatment strategy under which predictions are made. We extend measures of calibration, discrimination (c-index and cumulative/dynamic AUCt) and overall prediction error (Brier score) to allow assessment of counterfactual performance. The methods are evaluated using a simulation study, including scenarios in which the methods should detect poor performance. Applying our methods in the context of liver transplantation shows that our procedure allows quantification of the performance of predictions supporting crucial decisions on organ allocation.
27
Narita, Yusuke, Kyohei Okumura, Akihiro Shimizu e Kohei Yata. "Counterfactual Learning with General Data-Generating Policies". Proceedings of the AAAI Conference on Artificial Intelligence 37, n. 8 (26 giugno 2023): 9286–93. http://dx.doi.org/10.1609/aaai.v37i8.26113.
Abstract (sommario):
Off-policy evaluation (OPE) attempts to predict the performance of counterfactual policies using log data from a different policy. We extend its applicability by developing an OPE method for a class of both full support and deficient support logging policies in contextual-bandit settings. This class includes deterministic bandit (such as Upper Confidence Bound) as well as deterministic decision-making based on supervised and unsupervised learning. We prove that our method's prediction converges in probability to the true performance of a counterfactual policy as the sample size increases. We validate our method with experiments on partly and entirely deterministic logging policies. Finally, we apply it to evaluate coupon targeting policies by a major online platform and show how to improve the existing policy.
28
Altmeyer, Patrick, Mojtaba Farmanbar, Arie Van Deursen e Cynthia C. S. Liem. "Faithful Model Explanations through Energy-Constrained Conformal Counterfactuals". Proceedings of the AAAI Conference on Artificial Intelligence 38, n. 10 (24 marzo 2024): 10829–37. http://dx.doi.org/10.1609/aaai.v38i10.28956.
Abstract (sommario):
Counterfactual explanations offer an intuitive and straightforward way to explain black-box models and offer algorithmic recourse to individuals. To address the need for plausible explanations, existing work has primarily relied on surrogate models to learn how the input data is distributed. This effectively reallocates the task of learning realistic explanations for the data from the model itself to the surrogate. Consequently, the generated explanations may seem plausible to humans but need not necessarily describe the behaviour of the black-box model faithfully. We formalise this notion of faithfulness through the introduction of a tailored evaluation metric and propose a novel algorithmic framework for generating Energy-Constrained Conformal Counterfactuals that are only as plausible as the model permits. Through extensive empirical studies, we demonstrate that ECCCo reconciles the need for faithfulness and plausibility. In particular, we show that for models with gradient access, it is possible to achieve state-of-the-art performance without the need for surrogate models. To do so, our framework relies solely on properties defining the black-box model itself by leveraging recent advances in energy-based modelling and conformal prediction. To our knowledge, this is the first venture in this direction for generating faithful counterfactual explanations. Thus, we anticipate that ECCCo can serve as a baseline for future research. We believe that our work opens avenues for researchers and practitioners seeking tools to better distinguish trustworthy from unreliable models.
29
Admassu, Tsehay. "Evaluation of Local Interpretable Model-Agnostic Explanation and Shapley Additive Explanation for Chronic Heart Disease Detection". Proceedings of Engineering and Technology Innovation 23 (1 gennaio 2023): 48–59. http://dx.doi.org/10.46604/peti.2023.10101.
Abstract (sommario):
This study aims to investigate the effectiveness of local interpretable model-agnostic explanation (LIME) and Shapley additive explanation (SHAP) approaches for chronic heart disease detection. The efficiency of LIME and SHAP are evaluated by analyzing the diagnostic results of the XGBoost model and the stability and quality of counterfactual explanations. Firstly, 1025 heart disease samples are collected from the University of California Irvine. Then, the performance of LIME and SHAP is compared by using the XGBoost model with various measures, such as consistency and proximity. Finally, Python 3.7 programming language with Jupyter Notebook integrated development environment is used for simulation. The simulation result shows that the XGBoost model achieves 99.79% accuracy, indicating that the counterfactual explanation of the XGBoost model describes the smallest changes in the feature values for changing the diagnosis outcome to the predefined output.
30
Kopečná, Vědunka. "Counterfactual Impact Evaluation of the Project Internships for Young Job Seekers". Central European Journal of Public Policy 10, n. 2 (1 dicembre 2016): 48–66. http://dx.doi.org/10.1515/cejpp-2016-0026.
Abstract (sommario):
Abstract The growing youth unemployment across Europe raises the need to take appropriate measures. One of steps taken towards decreasing it by the European Union has been the program Youth Guarantee, implemented by a number of member states. Despite the relatively lower youth unemployment, the Czech Republic has implemented this program as well, and supported the realization of the project Internships for Young Job Seekers, whose aim was to ease the transition for students from schools to the labour market thanks to internships in companies. The effects which internship related project bring for their participants have been evaluated in other EU countries, mainly in Germany, but also in Sweden or France. However, evidence about internships’ effectiveness has been missing for the Czech Republic, and this paper fills this existing knowledge gap with the use of counterfactual impact evaluation methods. In the paper, we have focused on examining the impacts of internships on personal income and economic status of trainees by using the propensity score matching, difference-in-differences estimation and two complementary methods – ordinary least squares and multinomial logit. The results confirmed a positive impact of internships on treated project participants regarding both outcome variables, and thus, are consistent with the majority of literature in the field.
31
Lopez Buenache, German. "Monetary policy evaluation. A counterfactual analysis based on dynamic factor models". Applied Economics Letters 24, n. 7 (8 luglio 2016): 460–66. http://dx.doi.org/10.1080/13504851.2016.1203053.
32
Cirkovic, Milan. "Counterfactuals and unphysical ceteris paribus: An explanatory fallacy". Filozofija i drustvo 24, n. 4 (2013): 143–60. http://dx.doi.org/10.2298/fid1304143c.
Abstract (sommario):
I reconsider a type of counterfactual argument often used in historical sciences on a recent widely discussed example of the so-called ?rare Earth? hypothesis in planetary sciences and astrobiology. The argument is based on the alleged ?rarity? of some crucial ingredient for the planetary habitability, which is, in Earth?s case, provided by contingent evolutionary development. For instance, the claim that a contingent fact of history which has created planet Jupiter enables shielding of Earth from most dangerous impact catastrophes, thus increasing Earth?s habitability, leads often to the conclusion that such state-of-affairs must be rare in the Galaxy. I argue that this reasoning is deeply flawed, for several closely related reasons. In addition, the relevance of the philosophical problem of transworld identity for this kind of historical reasoning in science is put forward. This highlights many explanatory problems one faces when using historical counterfactuals in study of complex, nonlinear dynamical systems - and bolsters the relevance of philosophy for evaluation of scientific explanatory claims.
33
Oosterhuis, Harrie. "Learning from user interactions with rankings". ACM SIGIR Forum 54, n. 2 (dicembre 2020): 1–2. http://dx.doi.org/10.1145/3483382.3483402.
Abstract (sommario):
Ranking systems form the basis for online search engines and recommendation services. They process large collections of items, for instance web pages or e-commerce products, and present the user with a small ordered selection. The goal of a ranking system is to help a user find the items they are looking for with the least amount of effort. Thus the rankings they produce should place the most relevant or preferred items at the top of the ranking. Learning to rank is a field within machine learning that covers methods which optimize ranking systems w.r.t. this goal. Traditional supervised learning to rank methods utilize expert-judgements to evaluate and learn, however, in many situations such judgements are impossible or infeasible to obtain. As a solution, methods have been introduced that perform learning to rank based on user clicks instead. The difficulty with clicks is that they are not only affected by user preferences, but also by what rankings were displayed. Therefore, these methods have to prevent being biased by other factors than user preference. This thesis concerns learning to rank methods based on user clicks and specifically aims to unify the different families of these methods. The first part of the thesis consists of three chapters that look at online learning to rank algorithms which learn by directly interacting with users. Its first chapter considers large scale evaluation and shows existing methods do not guarantee correctness and user experience, we then introduce a novel method that can guarantee both. The second chapter proposes a novel pairwise method for learning from clicks that contrasts with the previous prevalent dueling-bandit methods. Our experiments show that our pairwise method greatly outperforms the dueling-bandit approach. The third chapter further confirms these findings in an extensive experimental comparison, furthermore, we also show that the theory behind the dueling-bandit approach is unsound w.r.t. deterministic ranking systems. The second part of the thesis consists of four chapters that look at counterfactual learning to rank algorithms which learn from historically logged click data. Its first chapter takes the existing approach and makes it applicable to top- k settings where not all items can be displayed at once. It also shows that state-of-the-art supervised learning to rank methods can be applied in the counterfactual scenario. The second chapter introduces a method that combines the robust generalization of feature-based models with the high-performance specialization of tabular models. The third chapter looks at evaluation and introduces a method for finding the optimal logging policy that collects click data in a way that minimizes the variance of estimated ranking metrics. By applying this method during the gathering of clicks, one can turn counterfactual evaluation into online evaluation. The fourth chapter proposes a novel counterfactual estimator that considers the possibility that the logging policy has been updated during the gathering of click data. As a result, it can learn much more efficiently when deployed in an online scenario where interventions can take place. The resulting approach is thus both online and counterfactual, our experimental results show that its performance matches the state-of-the-art in both the online and the counterfactual scenario. As a whole, the second part of this thesis proposes a framework that bridges many gaps between areas of online, counterfactual, and supervised learning to rank. It has taken approaches, previously considered independent, and unified them into a single methodology for widely applicable and effective learning to rank from user clicks. Awarded by: University of Amsterdam, Amsterdam, The Netherlands. Supervised by: Maarten de Rijke. Available at: https://hdl.handle.net/11245.1/8ff3aa38-97fb-4d2a-8127-a29a03af4d5c.
34
Cunha-e-Sá, Maria A., Rita Freitas, Luis Catela Nunes e Vladimir Otrachshenko. "On nature’s shoulders". Tourism Economics 24, n. 4 (20 settembre 2017): 369–85. http://dx.doi.org/10.1177/1354816617731195.
Abstract (sommario):
The use of counterfactual methods in the evaluation of policy interventions has been accepted today as the best approach in the estimation of a program’s performance. However, the simplest evaluations are often quite demanding in terms of the resources and the time needed to be implemented. In this article, we study the economic impact of a tourism media campaign launched in Nazaré, an old fishing community on the west coast of Portugal, to make big waves visible to the world. The campaign provided the required “informational media infrastructure” that created the public awareness necessary to boost tourism in the region. To measure the economic impact of that campaign on the local economy, we show how a counterfactual analysis can be implemented using regional statistical data on domestic and international tourist arrivals. We show how the method can be adapted to account for the presence of potential spillover effects that may have occurred, as neighboring municipalities could also have been affected by the intervention. We further compare the estimated impact on revenues with the costs incurred by the local municipality in the marketing campaign. Based on our empirical findings, we discuss policy implications to the municipalities in the region.
35
Kunimi, Takara, e Hajime Seya. "Identification of the geographical extent of an area benefiting from a transportation project: A generalized synthetic control". Journal of Transport and Land Use 14, n. 1 (5 gennaio 2021): 25–45. http://dx.doi.org/10.5198/jtlu.2021.1784.
Abstract (sommario):
In evaluating the benefits of an infrastructure project, it is essential to consider who is benefiting from the project and where benefits are located. However, there is no established way to accurately determine the latter. To fill this methodological gap, this study proposes an approach for the ex-post identification of the geographical extent of an area benefiting from a transportation project based on a generalized synthetic control method. Specifically, it allows comparing multiple treatment units with their counterfactuals in a single run—changes in land prices (actual outcome) at each treated site are compared to the counterfactual outcome, and the individual (i.e., unit-level) treatment effect on the treated site is then estimated. This approach is empirically applied to a large-scale Japanese heavy railway, the Tsukuba Express line project. Our approach enables the detection of 1) the complicated spatial shape of benefit incidence; 2) negative spillovers; and 3) the increase in options (train routes), typically not considered in a benefit evaluation system based on the hedonic approach, but which can be capitalized into land prices.
36
D’Alberto, Riccardo, Matteo Zavalloni, Meri Raggi e Davide Viaggi. "AES Impact Evaluation With Integrated Farm Data: Combining Statistical Matching and Propensity Score Matching". Sustainability 10, n. 11 (21 novembre 2018): 4320. http://dx.doi.org/10.3390/su10114320.
Abstract (sommario):
A large share of the Common Agricultural Policy (CAP) is allocated to agri-environmental schemes (AESs), whose goal is to foster the provision of a wide range of environmental public goods. Despite this effort, little is known on the actual environmental and economic impact of the AESs, due to the non-experimental conditions of the assessment exercise and several data availability issues. The main objective of the paper is to explore the feasibility of combining the non-parametric statistical matching (SM) method and propensity score matching (PSM) counterfactual approach analysis and to test its usefulness and practicability on a case study represented by selected impacts of the AESs in Emilia-Romagna. The work hints at the potentialities of the combined use of SM and PSM as well as of the systematic collection of additional information to be included in EU-financed project surveys in order to enrich and complete data collected in the official statistics. The results show that the combination of the two methods enables us to enlarge and deepen the scope of counterfactual analysis applied to AESs. In a specific case study, AESs seem to reduce the amount of rent-in land and decrease the crop mix diversity.
37
Benga, Elita, Juris Hāzners e Zaiga Miķelsone. "DISPLACEMENT EFFECTS OF LATVIAN RURAL DEVELOPMENT PROGRAMME 2007-2013". Environment. Technology. Resources. Proceedings of the International Scientific and Practical Conference 1 (15 giugno 2017): 28. http://dx.doi.org/10.17770/etr2017vol1.2662.
Abstract (sommario):
Periodic evaluation of EU Member States Rural Development Programme (RDP) specific policy interventions is considered crucial in policy development. The main reasons for the evaluation of specific policy interventions are the assessment of a programme’s impact, the improvement of programme management and administration, identification of necessary improvements in the delivery of interventions and meeting the accountability. The core question to be answered in programme evaluation is whether the stated objectives are accomplished by particular intervention (support or „treatment” provided to programme participants). The main problem in the process of evaluation is the assessment of the counterfactual outcome by modelling the situation where treatment is absent. The counterfactual outcome has to be estimated by statistical methods as it is usually not observed. General equilibrium effects occur when a programme affects units other than its participants. The most important possible impacts are the substitution effect and the displacement effect. Displacement effects are unplanned and indirect. They usually play a more important role in the evaluation at the programme level than in the evaluation of RDP individual measures. Displacement effect is the programme effect that occurs in a programme area at expense of another area. It takes place if farms located in one geographical area, which is not a subject to RD support, becomes adversely affected by a support provided to farms located in another geographically area. The existing study provides an assessment of the displacement effects on the employment in unsupported units at the programme level after the net effects on the employment calculated at the measure level are aggregated over the entire programme.
38
Sun, Yuewen, Erli Wang, Biwei Huang, Chaochao Lu, Lu Feng, Changyin Sun e Kun Zhang. "ACAMDA: Improving Data Efficiency in Reinforcement Learning through Guided Counterfactual Data Augmentation". Proceedings of the AAAI Conference on Artificial Intelligence 38, n. 14 (24 marzo 2024): 15193–201. http://dx.doi.org/10.1609/aaai.v38i14.29442.
Abstract (sommario):
Data augmentation plays a crucial role in improving the data efficiency of reinforcement learning (RL). However, the generation of high-quality augmented data remains a significant challenge. To overcome this, we introduce ACAMDA (Adversarial Causal Modeling for Data Augmentation), a novel framework that integrates two causality-based tasks: causal structure recovery and counterfactual estimation. The unique aspect of ACAMDA lies in its ability to recover temporal causal relationships from limited non-expert datasets. The identification of the sequential cause-and-effect allows the creation of realistic yet unobserved scenarios. We utilize this characteristic to generate guided counterfactual datasets, which, in turn, substantially reduces the need for extensive data collection. By simulating various state-action pairs under hypothetical actions, ACAMDA enriches the training dataset for diverse and heterogeneous conditions. Our experimental evaluation shows that ACAMDA outperforms existing methods, particularly when applied to novel and unseen domains.
39
Latour, Chiara, Franco Peracchi e Giancarlo Spagnolo. "Assessing alternative indicators for Covid-19 policy evaluation, with a counterfactual for Sweden". PLOS ONE 17, n. 3 (16 marzo 2022): e0264769. http://dx.doi.org/10.1371/journal.pone.0264769.
Abstract (sommario):
Using the synthetic control method, we construct counterfactuals for what would have happened if Sweden had imposed a lockdown during the first wave of the COVID-19 epidemic. We consider eight different indicators, including a novel one that we construct by adjusting recorded daily COVID-19 deaths to account for weakly excess mortality. Correcting for data problems and re-optimizing the synthetic control for each indicator, we find that a lockdown would have had sizable effects within one week. The much longer delay estimated by two previous studies focusing on the number of positives cases is mainly driven by the extremely low testing frequency that prevailed in Sweden in the first months of the epidemic. This result appears relevant for choosing the timing of future lockdowns and highlights the importance of looking at several indicators to derive robust conclusions. We also find that our novel indicator is effective in correcting errors in the COVID-19 deaths series and that the quantitative effects of the lockdown are stronger than previously estimated.
40
Markman, Keith D., Matthew N. McMullen e Ronald A. Elizaga. "Counterfactual thinking, persistence, and performance: A test of the Reflection and Evaluation Model". Journal of Experimental Social Psychology 44, n. 2 (marzo 2008): 421–28. http://dx.doi.org/10.1016/j.jesp.2007.01.001.
41
Tian, Meng, Tongping Li, Shuwang Yang, Yiwei Wang e Shuke Fu. "The Impact of High-Speed Rail on the Service-Sector Agglomeration in China". Sustainability 11, n. 7 (10 aprile 2019): 2128. http://dx.doi.org/10.3390/su11072128.
Abstract (sommario):
High-speed rail (HSR) can potentially influence various economic activities across space. Estimating the impacts of HSR on service-sector agglomeration (SSA) was proven to be difficult but meaningful. In this paper, prefectural-level data from 1998 to 2016 and a panel data program evaluation method are employed to evaluate the effect of the Wuhan–Guangzhou HSR (WGHSR) on the SSA along the route. In this way, we construct hypothetical counterfactuals for SSA index of the WGHSR cities in the absence of the HSR projects using the SSA index in selected non-HSR cities. By comparing the counterfactual index and the actual index, the evaluation of the WGHSR’s impact on the SSA along the route can be obtained. The results show that: (a) the WGHSR has increased the SSA by 9.44% on average for cities along the WGHSR, and (b) the impacts of the WGHSR on the SSA are heterogeneous. The WGHSR has brought about both spillover effect and “siphon” effect. In addition, whether the HSR influences and how much it could influence the SSA requires other supplementary conditions to be met. As a national strategy in China, the construction plan of HSR should fully consider its heterogeneity of impact on regional development. Policies should be formulated to drive the HSR’s spillover effect to promote regional sustainable development.
42
Borgonovo, Claudia. "Modales ambiguos". Revue Romane / Langue et littérature. International Journal of Romance Languages and Literatures 46, n. 2 (31 dicembre 2011): 202–21. http://dx.doi.org/10.1075/rro.46.2.02bor.
Abstract (sommario):
This article explores the interaction of Tense, Aspect and Modality in French, Italian and Spanish, languages in which Modals are inflected as main verbs. Imperfective modals are a-averidical, as modals are expected to be, but when they appear in a perfective tense, unexpected entailments and implicatures appear. For example, the following example is three-way ambiguous in Spanish; the corresponding example is two way-ambiguous in Italian and French: P. may have won, could have won, managed to win the race The three readings, epistemic, counterfactual and implicative, are derived from the alternative orderings of three heads, Tense, Aspect and Modal; any ordering in which Modal scopes over Tense is out on semantic grounds; these leaves three possible orderings which result in the three readings. In the epistemic construal, Modal has scope over Tense and Aspect, which are read on the infinitive. As a result, Modal Evaluation Time is at Utterance Time and the lower infinitive is Past and Perfective. There is no interaction between Modal and the other two heads and averidicality is the result. In the counterfactual reading,Tense scopes over Modal, which in turn scopes over Aspect: the result is a past Modal Evaluation; Perfective Aspect makes the interval in which verifying instances of the lower event are sought bounded, which contributes settledness. These are the crucial ingredients for counterfactuality. When Tense and Aspect are both read on the modal, the lower event is entailed and an implicative reading ensues. Italian and French do not have the counterfactual reading because their only surviving perfective past is a morphological perfect which, since it involves a resulting state, is incompatible with counterfactuality. Spanish perfecto also lacks this reading.
43
Li, Xiang, Hyukku Lee e Seung-Lin Hong. "Evaluation of the Policy Effects of Free Trade Agreements: New Evidence from the Korea-China FTA". Journal of Korea Trade 26, n. 6 (31 ottobre 2022): 41–60. http://dx.doi.org/10.35611/jkt.2022.26.6.41.
Abstract (sommario):
Purpose – The policy implications of free trade agreements have traditionally been a matter of debate among economists. The official signing of the Korea-China Free Trade Agreement provides economists with a quasi-natural experiment to analyze the FTA’s policy effects. This article aims to more accurately understand the impact of Korea’s FTA accession on the macro economy. Design/methodology – This study adopts the counterfactual method based on panel data to find common factors in the generation process of macro data to fit the counterfactual path, to accurately evaluate the effect of the macro policy. Findings – Our research results show that the signing of the Korea-China FTA has a relatively significant short-term positive effect on Korea's economic growth. On average, Korea’s real GDP growth rate has increased by 2.1%. This study finds evidence in support of FTA signing not having a significant impact on Korea’s GDP growth in the long run. Additionally, we evaluated the impact of the FTA on Korea’s imports and exports and found that it had a significant positive impact in the short term, but the trade effect of the FTA is significantly affected by the external macro-environment. Originality/value – First, this study uses macro panel data at the national level to examine the impact of the Korea-China FTA on Korea, and more accurately describes the policy effect of the FTA. Second, our empirical results show that the Korea-China FTA policy impact is subject to occasional changes in the external environment, such as the geopolitical conflict (crisis) between Korea and China, and the US-China trade war. Finally, the analysis shows that the short-term effect of FTA is significant but the long-term is uncertain, which provides empirical evidence for the debate on whether joining FTA can promote national economic growth.
44
Kazemi, Amirreza, e Martin Ester. "Adversarially Balanced Representation for Continuous Treatment Effect Estimation". Proceedings of the AAAI Conference on Artificial Intelligence 38, n. 12 (24 marzo 2024): 13085–93. http://dx.doi.org/10.1609/aaai.v38i12.29207.
Abstract (sommario):
Individual treatment effect (ITE) estimation requires adjusting for the covariate shift between populations with different treatments, and deep representation learning has shown great promise in learning a balanced representation of covariates. However the existing methods mostly consider the scenario of binary treatments. In this paper, we consider the more practical and challenging scenario in which the treatment is a continuous variable (e.g. dosage of a medication), and we address the two main challenges of this setup. We propose the adversarial counterfactual regression network (ACFR) that adversarially minimizes the representation imbalance in terms of KL divergence, and also maintains the impact of the treatment value on the outcome prediction by leveraging an attention mechanism. Theoretically we demonstrate that ACFR objective function is grounded in an upper bound on counterfactual outcome prediction error. Our experimental evaluation on semi-synthetic datasets demonstrates the empirical superiority of ACFR over a range of state-of-the-art methods.
45
Panfiluk, Eugenia. "Analysis of the effectiveness in the disbursement of the European Regional Development Fund for selected entities in the tourism economy". Ekonomia i Zarzadzanie 8, n. 4 (1 dicembre 2016): 39–49. http://dx.doi.org/10.1515/emj-2016-0031.
Abstract (sommario):
Abstract The aim of the article is to analyse the effectiveness of decision-making in the disbursement of funds from the ERDF for the selected tourism services. In the theoretical part of the article, the model for the assessment of the effectiveness of the ERDF management system was developed. This model is built based on the Propensity Score Matching (PSM) method, used to assess the effectiveness of community programmes. In the empirical part of the article, the verification of the model was carried out following the example of the tourism sector of the NUTS 2 region of Podlasie. The obtained results allow concluding that the developed deadweight assessment model can be used in practice for the evaluation of the efficiency of the ERDF management. The conducted research indicates that the method of Propensity Score Matching (PSM), using the Neyman-Rubin model based on counterfactual studies, allows assessing the decision-making effectiveness in the disbursement of funds from the ERDF. Creating counterfactual situations in the models of evaluation of infrastructural actions serves for the identification of the effect of an independent event. The effect of an independent event measures the negative, unintended consequences of the undertaken decisions. It informs about the loss of alternative possibilities of allocating funds for other purposes.
46
Porro, Giuseppe, e Valentina Salis. "Do local subsidies to firms create jobs? Counterfactual evaluation of an Italian regional experience". Papers in Regional Science 97, n. 4 (20 settembre 2017): 1039–56. http://dx.doi.org/10.1111/pirs.12317.
47
Joachims, Thorsten, Ben London, Yi Su, Adith Swaminathan e Lequn Wang. "Recommendations as Treatments". AI Magazine 42, n. 3 (20 novembre 2021): 19–30. http://dx.doi.org/10.1609/aimag.v42i3.18141.
Abstract (sommario):
In recent years, a new line of research has taken an interventional view of recommender systems, where recommendations are viewed as actions that the system takes to have a desired effect. This interventional view has led to the development of counterfactual inference techniques for evaluating and optimizing recommendation policies. This article explains how these techniques enable unbiased offline evaluation and learning despite biased data, and how they can inform considerations of fairness and equity in recommender systems.
48
Riddle Harding, Jennifer. "Evaluative stance and counterfactuals in language and literature". Language and Literature: International Journal of Stylistics 16, n. 3 (agosto 2007): 263–80. http://dx.doi.org/10.1177/0963947007079109.
Abstract (sommario):
This article argues that speakers often express attitudes not only toward events that have happened, but also toward counterfactual events; speakers communicate these attitudes by expressing an evaluative stance toward counterfactual scenarios. By analyzing examples from a variety of discourse situations, from conversation to canonical literature, the author demonstrates that counterfactuals and evaluations function jointly to produce rhetorical effects. The options for expressing evaluative stance are described in detail, as are the four configurations of focal scenario and evaluative stance that may arise in discourse. By considering the connection between evaluative stance and emotion, the author explains the rhetorical connection between counterfactuals and feelings of relief and regret. With this theoretical and methodological framework established, the article then moves to consider the role of counterfactuals and evaluative stance in literature. In literature, different speakers, including characters and narrators, may imagine and describe counterfactuals, which are scenarios not realized in the story and which are regarded as unrealized by the speaker who introduces them. These narrative speakers, as well as the implied author, may also adopt an evaluative stance toward counterfactuals that are introduced. Readers must juggle these contending representations and evaluations. Because counterfactual scenarios are often depictions of foreclosed possibilities, lost opportunities, and near misses linked to strong feelings of relief and regret, they are evocative elements of narrative that reward readers for their mental work with an enhanced appreciation for characters and textual themes. Counterfactuals also encourage readers to take a participatory role in the process of judgment.
49
Dindorf, Carlo, Oliver Ludwig, Steven Simon, Stephan Becker e Michael Fröhlich. "Machine Learning and Explainable Artificial Intelligence Using Counterfactual Explanations for Evaluating Posture Parameters". Bioengineering 10, n. 5 (24 aprile 2023): 511. http://dx.doi.org/10.3390/bioengineering10050511.
Abstract (sommario):
Postural deficits such as hyperlordosis (hollow back) or hyperkyphosis (hunchback) are relevant health issues. Diagnoses depend on the experience of the examiner and are, therefore, often subjective and prone to errors. Machine learning (ML) methods in combination with explainable artificial intelligence (XAI) tools have proven useful for providing an objective, data-based orientation. However, only a few works have considered posture parameters, leaving the potential for more human-friendly XAI interpretations still untouched. Therefore, the present work proposes an objective, data-driven ML system for medical decision support that enables especially human-friendly interpretations using counterfactual explanations (CFs). The posture data for 1151 subjects were recorded by means of stereophotogrammetry. An expert-based classification of the subjects regarding the presence of hyperlordosis or hyperkyphosis was initially performed. Using a Gaussian progress classifier, the models were trained and interpreted using CFs. The label errors were flagged and re-evaluated using confident learning. Very good classification performances for both hyperlordosis and hyperkyphosis were found, whereby the re-evaluation and correction of the test labels led to a significant improvement (MPRAUC = 0.97). A statistical evaluation showed that the CFs seemed to be plausible, in general. In the context of personalized medicine, the present study’s approach could be of importance for reducing diagnostic errors and thereby improving the individual adaptation of therapeutic measures. Likewise, it could be a basis for the development of apps for preventive posture assessment.
50
Svabova, Lucia, Eva Nahalkova Tesarova, Marek Durica e Lenka Strakova. "Evaluation of the impacts of the COVID-19 pandemic on the development of the unemployment rate in Slovakia: counterfactual before-after comparison". Equilibrium 16, n. 2 (30 giugno 2021): 261–84. http://dx.doi.org/10.24136/eq.2021.010.
Abstract (sommario):
Research background: The COVID-19 pandemic, which hit the world in the first quarter of 2020, has impacted almost every area of people's lives. Many states have introduced varying degrees of measures to prevent its spread. Most of these measures were, or still are, aimed at reducing or completely stopping the operation of shops and services, or in some cases, also the large manufacturing companies. However, as many companies have failed to cope with these restrictions, unemployment has risen in almost all EU countries. A similar situation was also observed in Slovakia, where the mentioned measures also had a significant impact on unemployment.
Purpose of the article: In this study, we deal with the quantification of the impact of a pandemic, or more precisely, anti-pandemic measures, on the development of the registered unemployment rate in Slovakia.
Methods: This quantification is based on the counterfactual method of before-after comparison, which is one of the most widely used methods in the field of impact assessments and brings very accurate results, based on real data. In the analysis, we use officially published data on the unemployment rate in Slovakia during the years 2013?2020 on a monthly basis. Such a long time series, using statistical methods of its decomposition and modelling of its trend, will allow predicting the development of the unemployment rate in Slovakia, assuming a counterfactual situation of no pandemic, and compare this development with the actual situation that occurred during 2020.
Findings & Value added: The study results indicate an increase in the unemployment rate in Slovakia during 2020 by 2?3% compared to the trend of its development, which would have occurred without a pandemic. Given the counterfactual method used, this difference can be described as the impact of the COVID-19 pandemic. The results of the study can be used in practice in the design and implementation of measures introduced to mitigate the impacts of the pandemic on unemployment and, in the long-term perspective, also to eliminate these effects as much as possible. It can also be used as a theoretical tool in conducting impact assessments, which have so far been carried out very rarely in Slovakia.