Collaborating Authors

Comparing Rewinding and Fine-tuning in Neural Network Pruning Machine Learning

Many neural network pruning algorithms proceed in three steps: train the network to completion, remove unwanted structure to compress the network, and retrain the remaining structure to recover lost accuracy. The standard retraining technique, fine-tuning, trains the unpruned weights from their final trained values using a small fixed learning rate. In this paper, we compare fine-tuning to alternative retraining techniques. Weight rewinding (as proposed by Frankle et al., (2019)), rewinds unpruned weights to their values from earlier in training and retrains them from there using the original training schedule. Learning rate rewinding (which we propose) trains the unpruned weights from their final values using the same learning rate schedule as weight rewinding. Both rewinding techniques outperform fine-tuning, forming the basis of a network-agnostic pruning algorithm that matches the accuracy and compression ratios of several more network-specific state-of-the-art techniques.

AI 'judge' can predict court verdicts with 79 per cent accuracy


Imagine this algorithm would then be refitted to replace the judge (or augment judges in a way that judges mostly rely on the tool to make a decision, which is already starting to happen [1]) and automatically put people in prison - with only 80% accuracy. Does that number still seem high now?

Adversarial Robustness Through Local Lipschitzness Machine Learning

A standard method for improving the robustness of neural networks is adversarial training, where the network is trained on adversarial examples that are close to the training inputs. This produces classifiers that are robust, but it often decreases clean accuracy. Prior work even posits that the tradeoff between robustness and accuracy may be inevitable. We investigate this tradeoff in more depth through the lens of local Lipschitzness. In many image datasets, the classes are separated in the sense that images with different labels are not extremely close in $\ell_\infty$ distance. Using this separation as a starting point, we argue that it is possible to achieve both accuracy and robustness by encouraging the classifier to be locally smooth around the data. More precisely, we consider classifiers that are obtained by rounding locally Lipschitz functions. Theoretically, we show that such classifiers exist for any dataset such that there is a positive distance between the support of different classes. Empirically, we compare the local Lipschitzness of classifiers trained by several methods. Our results show that having a small Lipschitz constant correlates with achieving high clean and robust accuracy, and therefore, the smoothness of the classifier is an important property to consider in the context of adversarial examples. Code available at .

Unlocking Fairness: a Trade-off Revisited

Neural Information Processing Systems

The prevailing wisdom is that a model's fairness and its accuracy are in tension with one another. However, there is a pernicious {\em modeling-evaluating dualism} bedeviling fair machine learning in which phenomena such as label bias are appropriately acknowledged as a source of unfairness when designing fair models, only to be tacitly abandoned when evaluating them. We investigate fairness and accuracy, but this time under a variety of controlled conditions in which we vary the amount and type of bias. We find, under reasonable assumptions, that the tension between fairness and accuracy is illusive, and vanishes as soon as we account for these phenomena during evaluation. Moreover, our results are consistent with an opposing conclusion: fairness and accuracy are sometimes in accord.