Convolution by Evolution: Differentiable Pattern Producing Networks — mentioned by leblancfg

Wed, 22 Jun 2016 15:28:32 GMT

First of all, thanks for the wonderful work, and I hope there's much more to come from your team! In fact I'm really pleased one of the authors came down to HN to comment. I think the scariest part of AI security is when the program itself becomes unfathomable. By that I mean, we can't just look at the source code and go "Ah! There's your problem". Now, your paper assumes a static reward function, but we can imagine the benefits of an AI that could dynamically change its reward function, or even its own source code. In fact, the most powerful tool I can think of to train a multi-purpose agent…

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

Convolution by Evolution: Differentiable Pattern Producing Networks — HN submission

Fri, 10 Jun 2016 15:35:22 GMT

Convolution by Evolution (from Google DeepMind)

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Convolution by Evolution: Differentiable Pattern Producing Networks · HackerPapers discussions

Convolution by Evolution: Differentiable Pattern Producing Networks — mentioned by leblancfg

Convolution by Evolution: Differentiable Pattern Producing Networks — HN submission