Skip to search formSkip to main contentSkip to account menu

Constrained Reinforcement Learning Has Zero Duality Gap

This work provides theoretical support to primal-dual approaches to reinforcement learning by establishing that despite its non-convexity, this problem has zero duality gap and can be solved exactly in the dual domain, where it becomes convex.
View PDF on arXiv (opens in a new tab)

Stability of Graph Scattering Transforms

This work extends scattering transforms to network data by using multiresolution graph wavelets, whose computation can be obtained by means of graph convolutions, and proves that the resulting graph scattering transforms are stable to metric perturbations of the underlying network.
View PDF on arXiv (opens in a new tab)

Stability Properties of Graph Neural Networks

This work proves that graph convolutions with integral Lipschitz filters, in combination with the frequency mixing effect of the corresponding nonlinearities, yields an architecture that is both stable to small changes in the underlying topology, and discriminative of information located at high frequencies.
View on IEEE (opens in a new tab)

On the Sample Complexity of Actor-Critic Method for Reinforcement Learning with Function Approximation

This work puts forth a new variant of actor-critic that employs Monte Carlo rollouts during the policy search updates, which results in controllable bias that depends on the number of critic evaluations, providing insight into the interplay between optimization and generalization in reinforcement learning.
View PDF on arXiv (opens in a new tab)

Safe Policies for Reinforcement Learning via Primal-Dual Methods

It is established that primal-dual algorithms are able to find policies that are safe and optimal, and an ergodic relaxation of the safe-learning problem is proposed.
View PDF on arXiv (opens in a new tab)

Network Newton-Part I: Algorithm and Convergence

Convergence to a point close to the optimal argument at a rate that is at least linear is proven and the existence of a tradeoff between convergence time and the distance to the optimum argument is shown.
View PDF on arXiv (opens in a new tab)

Graph Neural Networks for Decentralized Multi-Robot Path Planning

A combined model is proposed that automatically synthesizes local communication and decision-making policies for robots navigating in constrained workspaces that shows its capability to generalize to previously unseen cases.
View on IEEE (opens in a new tab)

Gated Graph Recurrent Neural Networks

Graph Recurrent Neural Networks are introduced as a general learning framework that achieves this goal by leveraging the notion of a recurrent hidden state together with graph signal processing (GSP) and the number of learnable parameters is independent of the length of the sequence and of the size of the graph, guaranteeing scalability.
View on IEEE (opens in a new tab)

EdgeNets: Edge Varying Graph Neural Networks

A general framework that unifies state-of-the-art graph neural networks (GNNs) through the concept of EdgeNet is put forth and it is shown that GATs are GCNNs on a graph that is learned from the features, which opens the doors to develop alternative attention mechanisms for improving discriminatory power.
View on IEEE (opens in a new tab)

Probably Approximately Correct Constrained Learning

It is proved that under mild conditions the empirical dual problem of constrained learning is also a PAC constrained learner that now leads to a practical constrained learning algorithm and is used to illustrate how constrained learning can address problems in fair and robust classification.
View PDF on arXiv (opens in a new tab)
...
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)