Editing a classifier by rewriting its prediction rules

• URL: http://arxiv.org/abs/2112.01008v1
• Date: Thu, 2 Dec 2021 06:40:37 GMT
• Title: Editing a classifier by rewriting its prediction rules
• Authors: Shibani Santurkar, Dimitris Tsipras, Mahalaxmi Elango, David Bau, Antonio Torralba, Aleksander Madry
• Abstract summary: We present a methodology for modifying the behavior of a classifier by directly rewriting its prediction rules. Our approach requires virtually no additional data collection and can be applied to a variety of settings, including adapting a model to new environments.
• Score: 133.5026383860842
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We present a methodology for modifying the behavior of a classifier by directly rewriting its prediction rules. Our approach requires virtually no additional data collection and can be applied to a variety of settings, including adapting a model to new environments, and modifying it to ignore spurious features. Our code is available at https://github.com/MadryLab/EditingClassifiers .

Related papers

• Generative Multi-modal Models are Good Class-Incremental Learners [51.5648732517187]
  We propose a novel generative multi-modal model (GMM) framework for class-incremental learning. Our approach directly generates labels for images using an adapted generative model. Under the Few-shot CIL setting, we have improved by at least 14% accuracy over all the current state-of-the-art methods with significantly less forgetting.
  arXiv  Detail & Related papers  (2024-03-27T09:21:07Z)
• Unsupervised Calibration through Prior Adaptation for Text Classification using Large Language Models [37.39843935632105]
  We propose an approach to adapt the prior class distribution to perform text classification tasks without the need for labelled samples. Results show that these methods outperform the un-adapted model for different number of training shots in the prompt.
  arXiv  Detail & Related papers  (2023-07-13T12:11:36Z)
• Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptors [53.819805242367345]
  We propose GRACE, a lifelong model editing method, which implements spot-fixes on streaming errors of a deployed model. GRACE writes new mappings into a pre-trained model's latent space, creating a discrete, local codebook of edits without altering model weights. Our experiments on T5, BERT, and GPT models show GRACE's state-of-the-art performance in making and retaining edits, while generalizing to unseen inputs.
  arXiv  Detail & Related papers  (2022-11-20T17:18:22Z)
• DiffusER: Discrete Diffusion via Edit-based Reconstruction [88.62707047517914]
  DiffusER is an edit-based generative model for text based on denoising diffusion models. It can rival autoregressive models on several tasks spanning machine translation, summarization, and style transfer. It can also perform other varieties of generation that standard autoregressive models are not well-suited for.
  arXiv  Detail & Related papers  (2022-10-30T16:55:23Z)
• Lightweight Conditional Model Extrapolation for Streaming Data under Class-Prior Shift [27.806085423595334]
  We introduce LIMES, a new method for learning with non-stationary streaming data. We learn a single set of model parameters from which a specific classifier for any specific data distribution is derived. Experiments on a set of exemplary tasks using Twitter data show that LIMES achieves higher accuracy than alternative approaches.
  arXiv  Detail & Related papers  (2022-06-10T15:19:52Z)
• Classifiers are Better Experts for Controllable Text Generation [63.17266060165098]
  We show that the proposed method significantly outperforms recent PPLM, GeDi, and DExperts on PPL and sentiment accuracy based on the external classifier of generated texts. The same time, it is also easier to implement and tune, and has significantly fewer restrictions and requirements.
  arXiv  Detail & Related papers  (2022-05-15T12:58:35Z)
• Unsupervised Learning of General-Purpose Embeddings for Code Changes [6.652641137999891]
  We propose an approach for obtaining embeddings of code changes during pre-training. We evaluate them on two different downstream tasks - applying changes to code and commit message generation. Our model outperforms the model that uses full edit sequences by 5.9 percentage points in accuracy.
  arXiv  Detail & Related papers  (2021-06-03T19:08:53Z)
• Rewriting a Deep Generative Model [56.91974064348137]
  We introduce a new problem setting: manipulation of specific rules encoded by a deep generative model. We propose a formulation in which the desired rule is changed by manipulating a layer of a deep network as a linear associative memory. We present a user interface to enable users to interactively change the rules of a generative model to achieve desired effects.
  arXiv  Detail & Related papers  (2020-07-30T17:58:16Z)

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.