DeepAI AI Chat
Log In Sign Up

SGD Learns the Conjugate Kernel Class of the Network

by   Amit Daniely, et al.

We show that the standard stochastic gradient decent (SGD) algorithm is guaranteed to learn, in polynomial time, a function that is competitive with the best function in the conjugate kernel space of the network, as defined in Daniely, Frostig and Singer. The result holds for log-depth networks from a rich family of architectures. To the best of our knowledge, it is the first polynomial-time guarantee for the standard neural network learning algorithm for networks of depth more that two. As corollaries, it follows that for neural networks of any depth between 2 and (n), SGD is guaranteed to learn, in polynomial time, constant degree polynomials with polynomially bounded coefficients. Likewise, it follows that SGD on large enough networks can learn any continuous function (not in polynomial time), complementing classical expressivity results.


page 1

page 2

page 3

page 4


Neural Networks Learning and Memorization with (almost) no Over-Parameterization

Many results in recent years established polynomial time learnability of...

Neural Networks Efficiently Learn Low-Dimensional Representations with SGD

We study the problem of training a two-layer neural network (NN) of arbi...

Recurrent Convolutional Neural Networks Learn Succinct Learning Algorithms

Neural Networks (NNs) struggle to efficiently learn certain problems, su...

Local SGD Optimizes Overparameterized Neural Networks in Polynomial Time

In this paper we prove that Local (S)GD (or FedAvg) can optimize two-lay...

The staircase property: How hierarchical structure can guide deep learning

This paper identifies a structural property of data distributions that e...

An Algorithm for Training Polynomial Networks

We consider deep neural networks, in which the output of each node is a ...

Contractility of continuous optimization

By introducing the concept of contractility, all the possible continuous...