1) Reinforcement Learning: these models leverage reinforcement learning action-reward duality
— TheSequence (@TheSequenceAI) November 16, 2021
2) Evolutionary Algorithms: population-based global optimizers for black-box functions
3) One-Shot Models: trains a single neural network during the search process
2/2 pic.twitter.com/yKPwUEgBqW
Tuesday, November 16, 2021
Neural architecture search (NAS): A ML technique for automating the creation of neural networks
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment