Regression Bugs Are In Your Model! Measuring, Reducing and Analyzing Regressions In NLP Model Updates

• URL: http://arxiv.org/abs/2105.03048v1
• Date: Fri, 7 May 2021 03:33:00 GMT
• Title: Regression Bugs Are In Your Model! Measuring, Reducing and Analyzing Regressions In NLP Model Updates
• Authors: Yuqing Xie, Yi-an Lai, Yuanjun Xiong, Yi Zhang, Stefano Soatto
• Abstract summary: This work focuses on quantifying, reducing and analyzing regression errors in the NLP model updates. We formulate the regression-free model updates into a constrained optimization problem. We empirically analyze how model ensemble reduces regression.
• Score: 68.09049111171862
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Behavior of deep neural networks can be inconsistent between different versions. Regressions during model update are a common cause of concern that often over-weigh the benefits in accuracy or efficiency gain. This work focuses on quantifying, reducing and analyzing regression errors in the NLP model updates. Using negative flip rate as regression measure, we show that regression has a prevalent presence across tasks in the GLUE benchmark. We formulate the regression-free model updates into a constrained optimization problem, and further reduce it into a relaxed form which can be approximately optimized through knowledge distillation training method. We empirically analyze how model ensemble reduces regression. Finally, we conduct CheckList behavioral testing to understand the distribution of regressions across linguistic phenomena, and the efficacy of ensemble and distillation methods.

Related papers

• Streaming Active Learning for Regression Problems Using Regression via Classification [12.572218568705376]
  We propose to use the regression-via-classification framework for streaming active learning for regression. Regression-via-classification transforms regression problems into classification problems so that streaming active learning methods can be applied directly to regression problems.
  arXiv  Detail & Related papers  (2023-09-02T20:24:24Z)
• Engression: Extrapolation through the Lens of Distributional Regression [2.519266955671697]
  We propose a neural network-based distributional regression methodology called engression' An engression model is generative in the sense that we can sample from the fitted conditional distribution and is also suitable for high-dimensional outcomes. We show that engression can successfully perform extrapolation under some assumptions such as monotonicity, whereas traditional regression approaches such as least-squares or quantile regression fall short under the same assumptions.
  arXiv  Detail & Related papers  (2023-07-03T08:19:00Z)
• Theoretical Characterization of the Generalization Performance of Overfitted Meta-Learning [70.52689048213398]
  This paper studies the performance of overfitted meta-learning under a linear regression model with Gaussian features. We find new and interesting properties that do not exist in single-task linear regression. Our analysis suggests that benign overfitting is more significant and easier to observe when the noise and the diversity/fluctuation of the ground truth of each training task are large.
  arXiv  Detail & Related papers  (2023-04-09T20:36:13Z)
• Improving Prediction Backward-Compatiblility in NLP Model Upgrade with Gated Fusion [8.173078054056337]
  When upgrading neural models to a newer version, new errors that were not encountered in the legacy version can be introduced, known as regression errors. We propose a novel method, Gated Fusion, that promotes backward compatibility via learning to mix predictions between old and new models.
  arXiv  Detail & Related papers  (2023-02-04T03:40:35Z)
• ResMem: Learn what you can and memorize the rest [79.19649788662511]
  We propose the residual-memorization (ResMem) algorithm to augment an existing prediction model. By construction, ResMem can explicitly memorize the training labels. We show that ResMem consistently improves the test set generalization of the original prediction model.
  arXiv  Detail & Related papers  (2023-02-03T07:12:55Z)
• Measuring and Reducing Model Update Regression in Structured Prediction for NLP [31.86240946966003]
  backward compatibility requires that the new model does not regress on cases that were correctly handled by its predecessor. This work studies model update regression in structured prediction tasks. We propose a simple and effective method, Backward-Congruent Re-ranking (BCR), by taking into account the characteristics of structured output.
  arXiv  Detail & Related papers  (2022-02-07T07:04:54Z)
• Human Pose Regression with Residual Log-likelihood Estimation [48.30425850653223]
  We propose a novel regression paradigm with Residual Log-likelihood Estimation (RLE) to capture the underlying output distribution. RLE learns the change of the distribution instead of the unreferenced underlying distribution to facilitate the training process. Compared to the conventional regression paradigm, regression with RLE bring 12.4 mAP improvement on MSCOCO without any test-time overhead.
  arXiv  Detail & Related papers  (2021-07-23T15:06:31Z)
• Flexible Model Aggregation for Quantile Regression [92.63075261170302]
  Quantile regression is a fundamental problem in statistical learning motivated by a need to quantify uncertainty in predictions. We investigate methods for aggregating any number of conditional quantile models. All of the models we consider in this paper can be fit using modern deep learning toolkits.
  arXiv  Detail & Related papers  (2021-02-26T23:21:16Z)
• Positive-Congruent Training: Towards Regression-Free Model Updates [87.25247195148187]
  In image classification, sample-wise inconsistencies appear as "negative flips" A new model incorrectly predicts the output for a test sample that was correctly classified by the old (reference) model. We propose a simple approach for PC training, Focal Distillation, which enforces congruence with the reference model.
  arXiv  Detail & Related papers  (2020-11-18T09:00:44Z)
• A Locally Adaptive Interpretable Regression [7.4267694612331905]
  Linear regression is one of the most interpretable prediction models. In this work, we introduce a locally adaptive interpretable regression (LoAIR) Our model achieves comparable or better predictive performance than the other state-of-the-art baselines.
  arXiv  Detail & Related papers  (2020-05-07T09:26:14Z)

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.