Theoretical Analysis of Weak-to-Strong Generalization

• URL: http://arxiv.org/abs/2405.16043v1
• Date: Sat, 25 May 2024 03:48:12 GMT
• Title: Theoretical Analysis of Weak-to-Strong Generalization
• Authors: Hunter Lang, David Sontag, Aravindan Vijayaraghavan,
• Abstract summary: We show that existing weak supervision theory fails to account for pseudolabel correction and coverage expansion. Our bounds capture the intuition that weak-to-strong generalization occurs when the strong model is unable to fit the mistakes of the weak teacher without incurring additional error.
• Score: 23.235671743867492
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Strong student models can learn from weaker teachers: when trained on the predictions of a weaker model, a strong pretrained student can learn to correct the weak model's errors and generalize to examples where the teacher is not confident, even when these examples are excluded from training. This enables learning from cheap, incomplete, and possibly incorrect label information, such as coarse logical rules or the generations of a language model. We show that existing weak supervision theory fails to account for both of these effects, which we call pseudolabel correction and coverage expansion, respectively. We give a new bound based on expansion properties of the data distribution and student hypothesis class that directly accounts for pseudolabel correction and coverage expansion. Our bounds capture the intuition that weak-to-strong generalization occurs when the strong model is unable to fit the mistakes of the weak teacher without incurring additional error. We show that these expansion properties can be checked from finite data and give empirical evidence that they hold in practice.

Related papers

• Wide Two-Layer Networks can Learn from Adversarial Perturbations [27.368408524000778]
  We theoretically explain the counterintuitive success of perturbation learning. We prove that adversarial perturbations contain sufficient class-specific features for networks to generalize from them.
  arXiv  Detail & Related papers  (2024-10-31T06:55:57Z)
• Toward Understanding In-context vs. In-weight Learning [50.24035812301655]
  We identify simplified distributional properties that give rise to the emergence and disappearance of in-context learning. We then extend the study to a full large language model, showing how fine-tuning on various collections of natural language prompts can elicit similar in-context and in-weight learning behaviour.
  arXiv  Detail & Related papers  (2024-10-30T14:09:00Z)
• Language Models Resist Alignment [8.4506780540122]
  Large language models (LLMs) may exhibit undesirable behaviors. Recent efforts have focused on aligning these models to prevent harmful generation. We show that fine-tuning process disproportionately undermines alignment compared to pre-training.
  arXiv  Detail & Related papers  (2024-06-10T10:03:16Z)
• Pre-training and Diagnosing Knowledge Base Completion Models [58.07183284468881]
  We introduce and analyze an approach to knowledge transfer from one collection of facts to another without the need for entity or relation matching. The main contribution is a method that can make use of large-scale pre-training on facts, which were collected from unstructured text. To understand the obtained pre-trained models better, we then introduce a novel dataset for the analysis of pre-trained models for Open Knowledge Base Completion.
  arXiv  Detail & Related papers  (2024-01-27T15:20:43Z)
• Classification and Adversarial examples in an Overparameterized Linear Model: A Signal Processing Perspective [10.515544361834241]
  State-of-the-art deep learning classifiers are highly susceptible to infinitesmal adversarial perturbations. We find that the learned model is susceptible to adversaries in an intermediate regime where classification generalizes but regression does not. Despite the adversarial susceptibility, we find that classification with these features can be easier than the more commonly studied "independent feature" models.
  arXiv  Detail & Related papers  (2021-09-27T17:35:42Z)
• Predicting Unreliable Predictions by Shattering a Neural Network [145.3823991041987]
  Piecewise linear neural networks can be split into subfunctions. Subfunctions have their own activation pattern, domain, and empirical error. Empirical error for the full network can be written as an expectation over subfunctions.
  arXiv  Detail & Related papers  (2021-06-15T18:34:41Z)
• Understanding Robustness in Teacher-Student Setting: A New Perspective [42.746182547068265]
  Adrial examples are machine learning models where bounded adversarial perturbation could mislead the models to make arbitrarily incorrect predictions. Extensive studies try to explain the existence of adversarial examples and provide ways to improve model robustness. Our studies could shed light on the future exploration about adversarial examples, and enhancing model robustness via principled data augmentation.
  arXiv  Detail & Related papers  (2021-02-25T20:54:24Z)
• Contrastive Learning Inverts the Data Generating Process [36.30995987986073]
  We prove that feedforward models trained with objectives belonging to the commonly used InfoNCE family learn to implicitly invert the underlying generative model of the observed data. Our theory highlights a fundamental connection between contrastive learning, generative modeling, and nonlinear independent component analysis.
  arXiv  Detail & Related papers  (2021-02-17T16:21:54Z)
• Don't Just Blame Over-parametrization for Over-confidence: Theoretical Analysis of Calibration in Binary Classification [58.03725169462616]
  We show theoretically that over-parametrization is not the only reason for over-confidence. We prove that logistic regression is inherently over-confident, in the realizable, under-parametrized setting. Perhaps surprisingly, we also show that over-confidence is not always the case.
  arXiv  Detail & Related papers  (2021-02-15T21:38:09Z)
• A Sober Look at the Unsupervised Learning of Disentangled Representations and their Evaluation [63.042651834453544]
  We show that the unsupervised learning of disentangled representations is impossible without inductive biases on both the models and the data. We observe that while the different methods successfully enforce properties "encouraged" by the corresponding losses, well-disentangled models seemingly cannot be identified without supervision. Our results suggest that future work on disentanglement learning should be explicit about the role of inductive biases and (implicit) supervision.
  arXiv  Detail & Related papers  (2020-10-27T10:17:15Z)
• Understanding Classifier Mistakes with Generative Models [88.20470690631372]
  Deep neural networks are effective on supervised learning tasks, but have been shown to be brittle. In this paper, we leverage generative models to identify and characterize instances where classifiers fail to generalize. Our approach is agnostic to class labels from the training set which makes it applicable to models trained in a semi-supervised way.
  arXiv  Detail & Related papers  (2020-10-05T22:13:21Z)

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.