The parallel repetition theorem states that for any two-prover game, with value 1 − (for, say, ≤ 1/2), the value of the game repeated in parallel n times is at most (1 − c)â...
It is shown that, for every strongly connected network in which every edge has capacity at least , linear time suffices to send flow from source vertices, each with a given supply,...
We propose a new approach to reinforcement learning which combines least squares function approximation with policy iteration. Our method is model-free and completely off policy. ...
A common problem in linear regression is that largely aberrant values can strongly influence the results. The least quartile difference (LQD) regression estimator is highly robus...
Thorsten Bernholt, Robin Nunkesser, Karen Schettli...
In a previous paper it was established that there are at least 27 non-isomorphic strongly regular (40, 12, 2, 4) graphs. Using a different and more efficient method we have re-inv...