Unifying and Verifying Mechanistic Interpretations: A Case Study with Group Operations

• URL: http://arxiv.org/abs/2410.07476v2
• Date: Fri, 11 Oct 2024 22:29:01 GMT
• Title: Unifying and Verifying Mechanistic Interpretations: A Case Study with Group Operations
• Authors: Wilson Wu, Louis Jaburi, Jacob Drori, Jason Gross,
• Abstract summary: A recent line of work in mechanistic interpretability has focused on reverse-engineering the computation performed by neural networks trained on the binary operation of finite groups. We investigate the internals of one-hidden-layer neural networks trained on this task, revealing previously unidentified structure. We produce a more complete description of such models that unifies the explanations of previous works.
• Score: 0.8305049591788082
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: A recent line of work in mechanistic interpretability has focused on reverse-engineering the computation performed by neural networks trained on the binary operation of finite groups. We investigate the internals of one-hidden-layer neural networks trained on this task, revealing previously unidentified structure and producing a more complete description of such models that unifies the explanations of previous works. Notably, these models approximate equivariance in each input argument. We verify that our explanation applies to a large fraction of networks trained on this task by translating it into a compact proof of model performance, a quantitative evaluation of model understanding. In particular, our explanation yields a guarantee of model accuracy that runs in 30% the time of brute force and gives a >=95% accuracy bound for 45% of the models we trained. We were unable to obtain nontrivial non-vacuous accuracy bounds using only explanations from previous works.

Related papers

• Improving Network Interpretability via Explanation Consistency Evaluation [56.14036428778861]
  We propose a framework that acquires more explainable activation heatmaps and simultaneously increase the model performance. Specifically, our framework introduces a new metric, i.e., explanation consistency, to reweight the training samples adaptively in model learning. Our framework then promotes the model learning by paying closer attention to those training samples with a high difference in explanations.
  arXiv  Detail & Related papers  (2024-08-08T17:20:08Z)
• SINDER: Repairing the Singular Defects of DINOv2 [61.98878352956125]
  Vision Transformer models trained on large-scale datasets often exhibit artifacts in the patch token they extract. We propose a novel fine-tuning smooth regularization that rectifies structural deficiencies using only a small dataset.
  arXiv  Detail & Related papers  (2024-07-23T20:34:23Z)
• Robust Nonparametric Hypothesis Testing to Understand Variability in Training Neural Networks [5.8490454659691355]
  We propose a new measure of closeness between classification models based on the output of the network before thresholding. Our measure is based on a robust hypothesis-testing framework and can be adapted to other quantities derived from trained models.
  arXiv  Detail & Related papers  (2023-10-01T01:44:35Z)
• VCNet: A self-explaining model for realistic counterfactual generation [52.77024349608834]
  Counterfactual explanation is a class of methods to make local explanations of machine learning decisions. We present VCNet-Variational Counter Net, a model architecture that combines a predictor and a counterfactual generator. We show that VCNet is able to both generate predictions, and to generate counterfactual explanations without having to solve another minimisation problem.
  arXiv  Detail & Related papers  (2022-12-21T08:45:32Z)
• Reconciliation of Pre-trained Models and Prototypical Neural Networks in Few-shot Named Entity Recognition [35.34238362639678]
  We propose a one-line-code normalization method to reconcile such a mismatch with empirical and theoretical grounds. Our work also provides an analytical viewpoint for addressing the general problems in few-shot name entity recognition.
  arXiv  Detail & Related papers  (2022-11-07T02:33:45Z)
• Harnessing the Power of Explanations for Incremental Training: A LIME-Based Approach [6.244905619201076]
  In this work, model explanations are fed back to the feed-forward training to help the model generalize better. The framework incorporates the custom weighted loss with Elastic Weight Consolidation (EWC) to maintain performance in sequential testing sets. The proposed custom training procedure results in a consistent enhancement of accuracy ranging from 0.5% to 1.5% throughout all phases of the incremental learning setup.
  arXiv  Detail & Related papers  (2022-11-02T18:16:17Z)
• SynBench: Task-Agnostic Benchmarking of Pretrained Representations using Synthetic Data [78.21197488065177]
  Recent success in fine-tuning large models, that are pretrained on broad data at scale, on downstream tasks has led to a significant paradigm shift in deep learning. This paper proposes a new task-agnostic framework, textitSynBench, to measure the quality of pretrained representations using synthetic data.
  arXiv  Detail & Related papers  (2022-10-06T15:25:00Z)
• Invariant Causal Mechanisms through Distribution Matching [86.07327840293894]
  In this work we provide a causal perspective and a new algorithm for learning invariant representations. Empirically we show that this algorithm works well on a diverse set of tasks and in particular we observe state-of-the-art performance on domain generalization.
  arXiv  Detail & Related papers  (2022-06-23T12:06:54Z)
• Building Reliable Explanations of Unreliable Neural Networks: Locally Smoothing Perspective of Model Interpretation [0.0]
  We present a novel method for reliably explaining the predictions of neural networks. Our method is built on top of the assumption of smooth landscape in a loss function of the model prediction.
  arXiv  Detail & Related papers  (2021-03-26T08:52:11Z)
• Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking [63.49779304362376]
  Graph neural networks (GNNs) have become a popular approach to integrating structural inductive biases into NLP models. We introduce a post-hoc method for interpreting the predictions of GNNs which identifies unnecessary edges. We show that we can drop a large proportion of edges without deteriorating the performance of the model.
  arXiv  Detail & Related papers  (2020-10-01T17:51:19Z)
• The Gaussian equivalence of generative models for learning with shallow neural networks [30.47878306277163]
  We study the performance of neural networks trained on data drawn from pre-trained generative models. We provide three strands of rigorous, analytical and numerical evidence corroborating this equivalence. These results open a viable path to the theoretical study of machine learning models with realistic data.
  arXiv  Detail & Related papers  (2020-06-25T21:20:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.