Skander Karkar

Skander Karkar, Ibrahim Ayed, Emmanuel de Bézenac et al.

Greedy layer-wise or module-wise training of neural networks is compelling in constrained and on-device settings where memory is limited, as it circumvents a number of problems of end-to-end back-propagation. However, it suffers from a stagnation problem, whereby early layers overfit and deeper layers stop increasing the test accuracy after a certain depth. We propose to solve this issue by introducing a module-wise regularization inspired by the minimizing movement scheme for gradient flows in distribution space. We call the method TRGL for Transport Regularized Greedy Learning and study it theoretically, proving that it leads to greedy modules that are regular and that progressively solve the task. Experimentally, we show improved accuracy of module-wise training of various architectures such as ResNets, Transformers and VGG, when our regularization is added, superior to that of other module-wise training methods and often to end-to-end training, with as much as 60% less memory usage.

Skander Karkar, Patrick Gallinari, Alain Rakotomamonjy

We propose a detector of adversarial samples that is based on the view of neural networks as discrete dynamic systems. The detector tells clean inputs from abnormal ones by comparing the discrete vector fields they follow through the layers. We also show that regularizing this vector field during training makes the network more regular on the data distribution's support, thus making the activations of clean inputs more distinguishable from those of abnormal ones. Experimentally, we compare our detector favorably to other detectors on seen and unseen attacks, and show that the regularization of the network's dynamics improves the performance of adversarial detectors that use the internal embeddings as inputs, while also improving test accuracy.

Skander Karkar, Ibrahim Ayed, Emmanuel de Bézenac et al.

End-to-end backpropagation has a few shortcomings: it requires loading the entire model during training, which can be impossible in constrained settings, and suffers from three locking problems (forward locking, update locking and backward locking), which prohibit training the layers in parallel. Solving layer-wise optimization problems can address these problems and has been used in on-device training of neural networks. We develop a layer-wise training method, particularly welladapted to ResNets, inspired by the minimizing movement scheme for gradient flows in distribution space. The method amounts to a kinetic energy regularization of each block that makes the blocks optimal transport maps and endows them with regularity. It works by alleviating the stagnation problem observed in layer-wise training, whereby greedily-trained early layers overfit and deeper layers stop increasing test accuracy after a certain depth. We show on classification tasks that the test accuracy of block-wise trained ResNets is improved when using our method, whether the blocks are trained sequentially or in parallel.

Skander Karkar, Ibrahim Ayed, Emmanuel de Bézenac et al.

Neural networks have been achieving high generalization performance on many tasks despite being highly over-parameterized. Since classical statistical learning theory struggles to explain this behavior, much effort has recently been focused on uncovering the mechanisms behind it, in the hope of developing a more adequate theoretical framework and having a better control over the trained models. In this work, we adopt an alternate perspective, viewing the neural network as a dynamical system displacing input particles over time. We conduct a series of experiments and, by analyzing the network's behavior through its displacements, we show the presence of a low kinetic energy displacement bias in the transport map of the network, and link this bias with generalization performance. From this observation, we reformulate the learning problem as follows: finding neural networks which solve the task while transporting the data as efficiently as possible. This offers a novel formulation of the learning problem which allows us to provide regularity results for the solution network, based on Optimal Transport theory. From a practical viewpoint, this allows us to propose a new learning algorithm, which automatically adapts to the complexity of the given task, and leads to networks with a high generalization ability even in low data regimes.