“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI

In recent years, many accurate decision support systems have been constructed as black boxes, that is as systems that hide their internal logic to the user. This lack of explanation constitutes both a practical and an ethical issue. The literature reports many approaches aimed at overcoming this crucial weakness, sometimes at the cost of sacrificing accuracy for interpretability. The applications in which black box decision systems can be used are various, and each approach is typically developed to provide a solution for a specific problem and, as a consequence, it explicitly or implicitly delineates its own definition of interpretability and explanation. The aim of this article is to provide a classification of the main problems addressed in the literature with respect to the notion of explanation and the type of black box system. Given a problem definition, a black box type, and a desired explanation, this survey should help the researcher to find the proposals more useful for his own work. The proposed classification of approaches to open black box models should also be useful for putting the many research open questions in perspective.

https://dl.acm.org/doi/pdf/10.1145/3236009

A Survey of Methods for Explaining Black Box Models

As machine learning systems become ubiquitous, there has been a surge of interest in interpretable machine learning: systems that provide explanation for their outputs. These explanations are often used to qualitatively assess other criteria such as safety or non-discrimination. However, despite the interest in interpretability, there is very little consensus on what interpretable machine learning is and how it should be measured. In this position paper, we first define interpretability and describe when interpretability is needed (and when it is not). Next, we suggest a taxonomy for rigorous evaluation and expose open questions towards a more rigorous science of interpretable machine learning.

Towards A Rigorous Science of Interpretable Machine Learning

Tree-based machine learning models such as random forests, decision trees and gradient boosted trees are popular nonlinear predictive models, yet comparatively little attention has been paid to explaining their predictions. Here we improve the interpretability of tree-based models through three main contributions. (1) A polynomial time algorithm to compute optimal explanations based on game theory. (2) A new type of explanation that directly measures local feature interaction effects. (3) A new set of tools for understanding global model structure based on combining many local explanations of each prediction. We apply these tools to three medical machine learning problems and show how combining many high-quality local explanations allows us to represent global structure while retaining local faithfulness to the original model. These tools enable us to (1) identify high-magnitude but low-frequency nonlinear mortality risk factors in the US population, (2) highlight distinct population subgroups with shared risk characteristics, (3) identify nonlinear interaction effects among risk factors for chronic kidney disease and (4) monitor a machine learning model deployed in a hospital by identifying which features are degrading the model’s performance over time. Given the popularity of tree-based machine learning models, these improvements to their interpretability have implications across a broad set of domains. Tree-based machine learning models are widely used in domains such as healthcare, finance and public services. The authors present an explanation method for trees that enables the computation of optimal local explanations for individual predictions, and demonstrate their method on three medical datasets.

/pdf/from-local-explanations-to-global-understanding-with-2a9bkpgseg.pdf

From Local Explanations to Global Understanding with Explainable AI for Trees.

Most people participate in meetings almost every day, multiple times a day. The study of meetings is important, but also challenging, as it requires an understanding of social signals and complex interpersonal dynamics. Our aim this work is to use a data-driven approach to the science of meetings. We provide tentative evidence that: i) it is possible to automatically detect when during the meeting a key decision is taking place, from analyzing only the local dialogue acts, ii) there are common patterns in the way social dialogue acts are interspersed throughout a meeting, iii) at the time key decisions are made, the amount of time left in the meeting can be predicted from the amount of time that has passed, iv) it is often possible to predict whether a proposal during a meeting will be accepted or rejected based entirely on the language (the set of persuasive words) used by the speaker.

Learning About Meetings

We aim to reduce the burden of programming and deploying autonomous systems to work in concert with people in time-critical domains such as military field operations and disaster response. Deployment plans for these operations are frequently negotiated on-the-fly by teams of human planners. A human operator then translates the agreed-upon plan into machine instructions for the robots. We present an algorithm that reduces this translation burden by inferring the final plan from a processed form of the human team's planning conversation. Our hybrid approach combines probabilistic generative modeling with logical plan validation used to compute a highly structured prior over possible plans, enabling us to overcome the challenge of performing inference over a large solution space with only a small amount of noisy data from the team planning session. We validate the algorithm through human subject experimentations and show that it is able to infer a human team's final plan with 86% accuracy on average. We also describe a robot demonstration in which two people plan and execute a first-response collaborative task with a PR2 robot. To the best of our knowledge, this is the first work to integrate a logical planning technique within a generative model to perform plan inference.

/pdf/inferring-team-task-plans-from-human-meetings-a-generative-34tewv1u86.pdf

Inferring team task plans from human meetings: a generative modeling approach with logic-based prior

Language models are known to learn a great quantity of factual information during pretraining, and recent work localizes this information to specific model weights like mid-layer MLP weights (Meng et al., 2022). In this paper, we find that we can change how a fact is stored in a model by editing weights that are in a different location than where existing methods suggest that the fact is stored. This is surprising because we would expect that localizing facts to specific parameters in models would tell us where to manipulate knowledge in models, and this assumption has motivated past work on model editing methods. Specifically, we show that localization conclusions from representation denoising (also known as Causal Tracing) do not provide any insight into which model MLP layer would be best to edit in order to override an existing stored fact with a new one. This finding raises questions about how past work relies on Causal Tracing to select which model layers to edit (Meng et al., 2022). Next, to better understand the discrepancy between representation denoising and weight editing, we develop several variants of the editing problem that appear more and more like representation denoising in their design and objective. Experiments show that, for one of our editing problems, editing performance does relate to localization results from representation denoising, but we find that which layer we edit is a far better predictor of performance. Our results suggest, counterintuitively, that better mechanistic understanding of how pretrained language models work may not always translate to insights about how to best change their behavior. Code is available at: https://github.com/google/belief-localization

Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models

Despite a sea of interpretability methods that can produce plausible explanations, the field has also empirically seen many failure cases of such methods. In light of these results, it remains unclear for practitioners how to use these methods and choose between them in a principled way. In this paper, we show that for moderately rich model classes (easily satisfied by neural networks), any feature attribution method that is complete and linear -- for example, Integrated Gradients and SHAP -- can provably fail to improve on random guessing for inferring model behaviour. Our results apply to common end-tasks such as characterizing local model behaviour, identifying spurious features, and algorithmic recourse. One takeaway from our work is the importance of concretely defining end-tasks: once such an end-task is defined, a simple and direct approach of repeated model evaluations can outperform many other complex feature attribution methods.

Impossibility Theorems for Feature Attribution

Research in both machine learning and psychology suggests that salient examples can help humans to interpret learning models. To this end, we take a novel look at black box interpretation of test predictions in terms of training examples. Our goal is to ask `which training examples are most responsible for a given set of predictions'? To answer this question, we make use of Fisher kernels as the defining feature embedding of each data point, combined with Sequential Bayesian Quadrature (SBQ) for efficient selection of examples. In contrast to prior work, our method is able to seamlessly handle any sized subset of test predictions in a principled way. We theoretically analyze our approach, providing novel convergence bounds for SBQ over discrete candidate atoms. Our approach recovers the application of influence functions for interpretability as a special case yielding novel insights from this connection. We also present applications of the proposed approach to three use cases: cleaning training data, fixing mislabeled examples and data summarization.

Been Kim

Papers

Learning About Meetings

Inferring team task plans from human meetings: a generative modeling approach with logic-based prior

Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models

Impossibility Theorems for Feature Attribution

Interpreting Black Box Predictions using Fisher Kernels