Greedy Pruning with Group Lasso Provably Generalizes for Matrix Sensing

Nived Rajaraman, Fnu Devvrit, Aryan Mokhtari, Kannan Ramchandran

Keywords: Greedy Pruning; Matrix Sensing; Lasso regularization

TL;DR: We provide the first rigorous theoretical insights on why the popular "greedy pruning+finetuning" pipeline leads to smaller models with good generalization, in the context of the overparameterized matrix sensing problem.

Abstract: Pruning schemes have been widely used in practice to reduce the complexity of trained models with a massive number of parameters. In fact, several practical studies have shown that if the pruned model is fine-tuned with some gradient-based updates it generalizes well to new samples. Although the above pipeline, which we refer to as pruning + fine-tuning, has been extremely successful in lowering the complexity of trained models, there is very little known about the theory behind this success. In this paper we address this issue by investigating the pruning + fine-tuning framework on the overparameterized matrix sensing problem with the ground truth denoted $U_\star \in \mathbb{R}^{d \times r}$ and the overparameterized model $U \in \mathbb{R}^{d \times k}$ with $k \gg r$. We study the approximate local minima of the mean square error, augmented with a smooth version of a group Lasso regularizer, $\sum_{i=1}^{k} \lVert Ue_i \rVert_2 $. In particular, we provably show that pruning all the columns below a certain explicit $\ell_2$-norm threshold results in a solution $U_{\text{prune}}$ which has the minimum number of columns $r$, yet close to the ground truth in training loss. Moreover, in the subsequent fine-tuning phase, gradient descent initialized at $U_{\text{prune}}$ converges at a linear rate to its limit. While our analysis provides insights into the role of regularization in pruning, we also show that running gradient descent in the absence of regularization results in models which {are not suitable for greedy pruning}, i.e., many columns could have their $\ell_2$ norm comparable to that of the maximum. Lastly, we show that our results also extend for the training and pruning of two-layer neural networks with quadratic activation functions. To the best of our knowledge, our results provide the first rigorous insights on why greedy pruning + fine-tuning leads to smaller models which also generalize well.

Submission Number: 6675