Implicit regularisation in stochastic gradient descent: from single-objective to two-player games.
Mihaela RoscaMarc Peter DeisenrothPublished in: CoRR (2023)
Keyphrases
- stochastic gradient descent
- two player games
- evaluation function
- loss function
- least squares
- matrix factorization
- step size
- random forests
- game tree
- model selection
- regularization parameter
- support vector machine
- multiple kernel learning
- temporal difference learning
- weight vector
- reinforcement learning
- feature selection
- support vector
- importance sampling
- random forest
- search algorithm
- convergence speed
- learning algorithm
- image restoration
- linear combination
- simulated annealing
- recommender systems
- image segmentation
- pairwise