🏢 CREST, ENSAE, IP Paris
The Value of Reward Lookahead in Reinforcement Learning
·1360 words·7 mins·
loading
·
loading
Reinforcement Learning
🏢 CREST, ENSAE, IP Paris
Reinforcement learning agents can achieve significantly higher rewards by using advance knowledge of future rewards; this paper mathematically analyzes this advantage by computing the worst-case perfo…
Statistical and Geometrical properties of the Kernel Kullback-Leibler divergence
·1547 words·8 mins·
loading
·
loading
AI Theory
Optimization
🏢 CREST, ENSAE, IP Paris
Regularized Kernel Kullback-Leibler divergence solves the original KKL’s disjoint support limitation, enabling comparison of any probability distributions with a closed-form solution and efficient gra…
Mirror and Preconditioned Gradient Descent in Wasserstein Space
·1610 words·8 mins·
loading
·
loading
AI Theory
Optimization
🏢 CREST, ENSAE, IP Paris
This paper presents novel mirror and preconditioned gradient descent algorithms for optimizing functionals over Wasserstein space, offering improved convergence and efficiency for various machine lear…