Increasing Performance And Sample Efficiency With Model-agnostic Interactive Feature Attributions
Model-agnostic feature attributions can provide local insights in complex ML models. If the explanation is correct, a domain expert can validate and trust the model's decision. However, if it contradicts the expert's knowledge, related work only corrects irrelevant features to improve the model. To allow for unlimited interaction, in this paper we provide model-agnostic implementations for two popular explanation methods (Occlusion and Shapley values) to enforce entirely different attributions in the complex model. For a particular set of samples, we use the corrected feature attributions to generate extra local data, which is used to retrain the model to have the right explanation for the samples. Through simulated and real data experiments on a variety of models we show how our proposed approach can significantly improve the model's performance only by augmenting its training dataset based on corrected explanations. Adding our interactive explanations to active learning settings increases the sample efficiency significantly and outperforms existing explanatory interactive strategies. Additionally we explore how a domain expert can provide feature attributions which are sufficiently correct to improve the model.
Code (0)
등록된 구현이 없습니다.
Tasks
Active LearningSimilar Papers 제목 키워드 기반
Agnostic Interactive Imitation Learning: New Theory and Practical Algorithms
We study interactive imitation learning, where a learner interactively queries a demonstrating expert for action annotations, aiming to learn a policy that has performance competitive with the expert, using as few annota…
continuous-controlContinuous ControlImitation LearningLearning versus Refutation in Noninteractive Local Differential Privacy
We study two basic statistical tasks in non-interactive local differential privacy (LDP): learning and refutation. Learning requires finding a concept that best fits an unknown target function (from labelled samples draw…
PAC learningImproving Interactive Diagnostic Ability of a Large Language Model Agent Through Clinical Experience Learning
Recent advances in large language models (LLMs) have shown promising results in medical diagnosis, with some studies indicating superior performance compared to human physicians in specific scenarios. However, the diagno…
DiagnosticLanguage ModelingLanguage ModellingLarge Language Model+1On the Power of Interactive Proofs for Learning
We continue the study of doubly-efficient proof systems for verifying agnostic PAC learning, for which we obtain the following results. - We construct an interactive protocol for learning the $t$ largest Fourier characte…
PAC learningInteractive Segmentation and Report Generation for CT Images
Automated CT report generation plays a crucial role in improving diagnostic accuracy and clinical workflow efficiency. However, existing methods lack interpretability and impede patient-clinician understanding, while the…
AttributeDiagnosticInteractive SegmentationLesion Segmentation+1