conftrace_

tree backup

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

multi-step lookahead (1) n-step return (1) retrace algorithm (1) gradient-based algorithm (1) convergence guarantee (1) convex-concave saddle-point (1) policy improvement (1) off-policy learning (1) value function approximation (1) policy iteration (1)

Papers

Beyond the One-Step Greedy Approach in Reinforcement Learning ICML 2018 Convergent Tree Backup and Retrace with Function Approximation ICML 2018