Résultats de recherche pour « Interpretability »

We introduce the Concept Bottleneck Large Language Model (CB-LLM), a pioneering approach to creating inherently interpretable Large Language Models (LLMs). Unlike traditional black-box LLMs that rely on post-hoc interpretation methods with limited neuron function insights, CB-LLM sets a new standard with its built-in interpretability, scalability, and ability to provide clear, accurate explanations. This innovation not only advances transparency in language models but also enhances their effectiveness. Our unique Automatic Concept Correction (ACC) strategy successfully narrows the performance gap with conventional black-box LLMs, positioning CB-LLM as a model that combines the high accuracy of traditional LLMs with the added benefit of clear interpretability -- a feature markedly absent in existing LLMs.

Decisions by Machine Learning (ML) models have become ubiquitous. Trusting these decisions requires understanding how algorithms take them. Hence interpretability methods for ML are an active focus of research. A central problem in this context is that both the quality of interpretability methods as well as trust in ML predictions are difficult to measure. Yet evaluations, comparisons and improvements of trust and interpretability require quantifiable measures. Here we propose a quantitative measure for the quality of interpretability methods. Based on that we derive a quantitative measure of trust in ML decisions. Building on previous work we propose to measure intuitive understanding of algorithmic decisions using the information transfer rate at which humans replicate ML model predictions. We provide empirical evidence from crowdsourcing experiments that the proposed metric robustly differentiates interpretability methods. The proposed metric also demonstrates the value of interpretability for ML assisted human decision making: in our experiments providing explanations more than doubled productivity in annotation tasks. However unbiased human judgement is critical for doctors, judges, policy makers and others. Here we derive a trust metric that identifies when human decisions are overly biased towards ML predictions. Our results complement existing qualitative work on trust and interpretability by quantifiable measures that can serve as objectives for further improving methods in this field of research.

Supervised machine learning models boast remarkable predictive capabilities. But can you trust your model? Will it work in deployment? What else can it tell you about the world? We want models to be not only good, but interpretable. And yet the task of interpretation appears underspecified. Papers provide diverse and sometimes non-overlapping motivations for interpretability, and offer myriad notions of what attributes render models interpretable. Despite this ambiguity, many papers proclaim interpretability axiomatically, absent further explanation. In this paper, we seek to refine the discourse on interpretability. First, we examine the motivations underlying interest in interpretability, finding them to be diverse and occasionally discordant. Then, we address model properties and techniques thought to confer interpretability, identifying transparency to humans and post-hoc explanations as competing notions. Throughout, we discuss the feasibility and desirability of different notions, and question the oft-made assertions that linear models are interpretable and that deep neural networks are not.

Ethical AI spans a gamut of considerations. Among these, the most popular ones, fairness and interpretability, have remained largely distinct in technical pursuits. We discuss and elucidate the differences between fairness and interpretability across a variety of dimensions. Further, we develop two principles-based frameworks towards developing ethical AI for the future that embrace aspects of both fairness and interpretability. First, interpretability for fairness proposes instantiating interpretability within the realm of fairness to develop a new breed of ethical AI. Second, fairness and interpretability initiates deliberations on bringing the best aspects of both together. We hope that these two frameworks will contribute to intensifying scholarly discussions on new frontiers of ethical AI that brings together fairness and interpretability.

Improving the interpretability of brain decoding approaches is of primary interest in many neuroimaging studies. Despite extensive studies of this type, at present, there is no formal definition for interpretability of brain decoding models. As a consequence, there is no quantitative measure for evaluating the interpretability of different brain decoding methods. In this paper, we present a simple definition for interpretability of linear brain decoding models. Then, we propose to combine the interpretability and the performance of the brain decoding into a new multi-objective criterion for model selection. Our preliminary results on the toy data show that optimizing the hyper-parameters of the regularized linear classifier based on the proposed criterion results in more informative linear models. The presented definition provides the theoretical background for quantitative evaluation of interpretability in linear brain decoding.