Log-linear Guardedness and its Implications

• URL: http://arxiv.org/abs/2210.10012v5
• Date: Fri, 10 May 2024 20:59:02 GMT
• Title: Log-linear Guardedness and its Implications
• Authors: Shauli Ravfogel, Yoav Goldberg, Ryan Cotterell,
• Abstract summary: Methods for erasing human-interpretable concepts from neural representations that assume linearity have been found to be tractable and useful. This work formally defines the notion of log-linear guardedness as the inability of an adversary to predict the concept directly from the representation. We show that, in the binary case, under certain assumptions, a downstream log-linear model cannot recover the erased concept.
• Score: 116.87322784046926
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Methods for erasing human-interpretable concepts from neural representations that assume linearity have been found to be tractable and useful. However, the impact of this removal on the behavior of downstream classifiers trained on the modified representations is not fully understood. In this work, we formally define the notion of log-linear guardedness as the inability of an adversary to predict the concept directly from the representation, and study its implications. We show that, in the binary case, under certain assumptions, a downstream log-linear model cannot recover the erased concept. However, we demonstrate that a multiclass log-linear model \emph{can} be constructed that indirectly recovers the concept in some cases, pointing to the inherent limitations of log-linear guardedness as a downstream bias mitigation technique. These findings shed light on the theoretical limitations of linear erasure methods and highlight the need for further research on the connections between intrinsic and extrinsic bias in neural models.

Related papers

• Cross-Entropy Is All You Need To Invert the Data Generating Process [29.94396019742267]
  Empirical phenomena suggest that supervised models can learn interpretable factors of variation in a linear fashion. Recent advances in self-supervised learning have shown that these methods can recover latent structures by inverting the data generating process. We prove that even in standard classification tasks, models learn representations of ground-truth factors of variation up to a linear transformation.
  arXiv  Detail & Related papers  (2024-10-29T09:03:57Z)
• Bayesian Inference for Consistent Predictions in Overparameterized Nonlinear Regression [0.0]
  This study explores the predictive properties of over parameterized nonlinear regression within the Bayesian framework. Posterior contraction is established for generalized linear and single-neuron models with Lipschitz continuous activation functions. The proposed method was validated via numerical simulations and a real data application.
  arXiv  Detail & Related papers  (2024-04-06T04:22:48Z)
• On the Dynamics Under the Unhinged Loss and Beyond [104.49565602940699]
  We introduce the unhinged loss, a concise loss function, that offers more mathematical opportunities to analyze closed-form dynamics. The unhinged loss allows for considering more practical techniques, such as time-vary learning rates and feature normalization.
  arXiv  Detail & Related papers  (2023-12-13T02:11:07Z)
• Kernelized Concept Erasure [108.65038124096907]
  We propose a kernelization of a linear minimax game for concept erasure. It is possible to prevent specific non-linear adversaries from predicting the concept. However, the protection does not transfer to different nonlinear adversaries.
  arXiv  Detail & Related papers  (2022-01-28T15:45:13Z)
• Linear Adversarial Concept Erasure [108.37226654006153]
  We formulate the problem of identifying and erasing a linear subspace that corresponds to a given concept. We show that the method is highly expressive, effectively mitigating bias in deep nonlinear classifiers while maintaining tractability and interpretability.
  arXiv  Detail & Related papers  (2022-01-28T13:00:17Z)
• Benign Overfitting in Adversarially Robust Linear Classification [91.42259226639837]
  "Benign overfitting", where classifiers memorize noisy training data yet still achieve a good generalization performance, has drawn great attention in the machine learning community. We show that benign overfitting indeed occurs in adversarial training, a principled approach to defend against adversarial examples.
  arXiv  Detail & Related papers  (2021-12-31T00:27:31Z)
• Relaxing the Constraints on Predictive Coding Models [62.997667081978825]
  Predictive coding is an influential theory of cortical function which posits that the principal computation the brain performs is the minimization of prediction errors. Standard implementations of the algorithm still involve potentially neurally implausible features such as identical forward and backward weights, backward nonlinear derivatives, and 1-1 error unit connectivity. In this paper, we show that these features are not integral to the algorithm and can be removed either directly or through learning additional sets of parameters with Hebbian update rules without noticeable harm to learning performance.
  arXiv  Detail & Related papers  (2020-10-02T15:21:37Z)
• Linear Disentangled Representations and Unsupervised Action Estimation [2.793095554369282]
  We show that linear disentangled representations are not generally present in standard VAE models. We propose a method to induce irreducible representations which forgoes the need for labelled action sequences.
  arXiv  Detail & Related papers  (2020-08-18T13:23:57Z)

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.