Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning · HackerPapers discussions

Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning — mentioned by tree_of_item

Thu, 26 Jul 2018 15:08:42 GMT

People keep saying it's "not feasible" to do deep learning on a CPU. Is that actually true? I'm thinking of papers like this one[0] from Uber. that make the claim that CPU training is very much feasible, in this case with a cluster of 10 machines, each with 72~ CPUs. There's a blog post about someone recreating the paper here[1]. Sure, you can't do it on ONE CPU, but the point is to have a cluster of CPUs. It's not the case that you're forced to use nVidia's proprietary stuff in order to do deep learning. [0]: https://arxiv.org/abs/1712.06567 [1]…

Paper and discussion history. A mention is not an endorsement.

Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning — mentioned by pjrule

Thu, 07 Jun 2018 00:20:44 GMT

As someone working on a reinforcement learning/neuroevolution problem right now, I find this to be extremely exciting. Fewer parameters, ceteris paribus, is always better—the fact that the experiments in this paper were run on one workstation, rather than on a massive farm of TPUs à la AlphaGo, implies quicker development iteration time and more accessibility to the average researcher. The staging of components in this paper (compressor/controller), where neuroevolution is only applied to a low-dimensional controller, reminds me of Ha and Schmidhuber's recent paper on world models (which is…

Paper and discussion history. A mention is not an endorsement.

Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning — mentioned by Matumio

Mon, 25 Dec 2017 15:46:34 GMT

Even random search (in the weight space) can outperform gradient-based deep reinforcement learning algorithms on some Atari games. Genetic algorithms are quite competitive. Paper: https://arxiv.org/abs/1712.06567v1

Paper and discussion history. A mention is not an endorsement.

Deep Neuroevolution: Genetic Algorithms Are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning — HN submission

Thu, 21 Dec 2017 15:41:27 GMT

Genetic Algorithms for Training Deep Neural Networks for Reinforcement Learning

124 points; 44 discussion replies on HN. Paper and source history. A submission is not an endorsement.