Robert Tjarko Lange

neural network pruning (1) lottery ticket hypothesis (1) bayesian inference (1) preference optimization (1) network pruning (1) reinforcement learning from human feedback (1) gradient descent (1) offline learning (1) exponential loss (1) logistic loss (1) amortized inference (1) inductive bia (1) loss function (1) sparse training (1) convex loss function (1) algorithm discovery (1) neural network initialization (1) evolution strategies (1) large language model (1) adaptive behavior (1)