Learning model-based planning from scratch · HackerPapers discussions

Learning model-based planning from scratch — mentioned by gwern

Wed, 09 Aug 2017 19:04:42 GMT

I think it is doable in under 5 years, but this critically depends on the resources invested by DM and other DL orgs. Deep RL is hugely demanding of computational resources to iterate your designs - for example, the first AlphaGo took something like 3 GPU-years to train it once (2 or 3 months parallelized); however, with much more iteration, DM was able to get Master's from-scratch training down to under 1 month. Now an AG researcher can iterate rapidly with small-scale hobbyist or researcher resources, but if they had had to do it all themselves, Ke Jie would still be waiting for a worthy…

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

Learning model-based planning from scratch — mentioned by gradstudent

Tue, 25 Jul 2017 01:27:42 GMT

I'm not a planning guy but I work in a closely related community so I'm a least somewhat familar with the area. Looking at the first paper (https://arxiv.org/pdf/1707.06170.pdf), it seems surprisingly shallow and light on details. So they have a learning system for continuous planning. So what? The AI Planning community has been doing this for ages with MDPs and POMDPs, solving problems where the planning domain has some discrete variables and some continuous variables. Here's a summary tutorial from Scott Sanner at ICAPS 2012: http://icaps12.icaps-conference.org/planningschool/slides-Sa...…

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