Modern neural networks often have great expressive power and can be trained to overfit the
training data, while still achieving a good test performance. This phenomenon is referred to …

Modern deep learning models with great expressive power can be trained to overfit the
training data but still generalize well. This phenomenon is referred to as benign overfitting …

The implicit bias towards solutions with favorable properties is believed to be a key reason
why neural networks trained by gradient-based optimization can generalize well. While the …

The challenge of overfitting, in which the model memorizes the training data and fails to
generalize to test data, has become increasingly significant in the training of large neural …

The``Neural Tangent Kernel''(NTK)(Jacot et al 2018), and its empirical variants have been
proposed as a proxy to capture certain behaviors of real neural networks. In this work, we …

Meta learning has demonstrated tremendous success in few-shot learning with limited
supervised data. In those settings, the meta model is usually overparameterized. While the …

It is by now well-established that modern over-parameterized models seem to elude the bias-
variance tradeoff and generalize well despite overfitting noise. Many recent works attempt to …

We bound the excess risk of interpolating deep linear networks trained using gradient flow.
In a setting previously used to establish risk bounds for the minimum ℓ2-norm interpolant, we …

Adversarial training is one important algorithm to achieve robust machine learning models.
However, numerous empirical results show a great performance degradation from clean …

Transformers have demonstrated great power in the recent development of large
foundational models. In particular, the Vision Transformer (ViT) has brought revolutionary …