52 bookmarks. First posted by nirum october 2017.
In this post I explain how evolution strategies (ES) work with the aid of a few visual examples. I try to keep the equations light, and I provide links to original articles if the reader wishes to understand more details. This is the first post in a series of articles, where I plan to show how to apply these algorithms to a range of tasks from MNIST, OpenAI Gym, Roboschool to PyBullet environments.ai ml algorithm
17 days ago by aristidb
RL is devoted to estimate this credit-assignment problem, and great progress has been made in recent years. However, credit assignment is still difficult when the reward signals are sparse. In the real world, rewards can be sparse and noisy. Sometimes we are given just a single reward, like a bonus check at the end of the year, and depending on our employer, it may be difficult to figure out exactly why it is so low. For these problems, rather than rely on a very noisy and possibly meaningless gradient estimate of the future to our policy, we might as well just ignore any gradient information, and attempt to use black-box optimisation techniques such as genetic algorithms (GA) or ES.evolutionary geneticalgorithms AI reinforcementlearning deeplearning visualization algorithms tweetit
october 2017 by sachaa
tags1117 111217 ai algorithm algorithms biomimetics by:hardmaru data-science deeplearning es evolution evolutionary-algorithms evolutionary evolutionstrategies feedly genetic geneticalgorithm geneticalgorithms geneticprogramming genetic_algorithms graphics ifttt infovis later learning machine-learning machine machinelearning machine_learning ml neuralnet neuralnets neuralnetwork optimisation optimization pocket programming programming_evolution reading reinforcement-learning reinforcementlearning statistics teaching theory ththlink tweetit twitter via:chl via:popular visualization