Papers › Limitations of Lazy Training of Two-layers Neural Networks

Limitations of Lazy Training of Two-layers Neural Networks

21 Jun 2019arXiv:1906.08899archive 2025-07-28

Behrooz Ghorbani, Song Mei, Theodor Misiakiewicz, Andrea Montanari

We study the supervised learning problem under either of the following two models: (1) Feature vectors xᵢ are d-dimensional Gaussians and responses are yᵢ = f_*(xᵢ) for f_* an unknown quadratic function; (2) Feature vectors xᵢ are distributed as a mixture of two d-dimensional centered Gaussians, and yᵢ's are the corresponding class labels. We use two-layers neural networks with quadratic activations, and compare three different learning regimes: the random features (RF) regime in which we only train the second-layer weights; the neural tangent (NT) regime in which we train a linearization of the neural network around its initialization; the fully trained neural network (NN) regime in which we train all the weights in the network. We prove that, even for the simple quadratic model of point (1), there is a potentially unbounded gap between the prediction risk achieved in these three training regimes, when the number of neurons is smaller than the ambient dimension. When the number of neurons is larger than the number of dimensions, the problem is significantly easier and both NT and NN learning achieve zero risk.

Code

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections