DeepCoder: Learning to Write Programs · HackerPapers discussions

DeepCoder: Learning to Write Programs — mentioned by darkmighty

Mon, 14 Aug 2017 14:50:25 GMT

The evolutionary pressure is just people or programs detecting certain strains (it would generate lots of strains randomly). I think the only real issue to plausibility here is I'm not sure brute force is enough to make enough plausible behavioral branches, at least with current computing power/internet bandwidth. A reasonably efficient self-modification mechanism (in terms of viable strain per transmission) is probably extremely large, I'd say at least 1GB. Not unlike deep learning systems, this would consist of a large functional composition of heuristics, codifying how to write code that…

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

DeepCoder: Learning to Write Programs — mentioned by EvgeniyZh

Fri, 24 Feb 2017 08:16:22 GMT

That's interesting that the paper was on arxiv since November ( https://arxiv.org/abs/1611.01989 ) and people payed attention just now.

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

DeepCoder: Learning to Write Programs — HN submission

Fri, 24 Feb 2017 03:09:53 GMT

DeepCoder: Learning to Write Programs

183 points; 54 discussion replies on HN. Paper and source history. A submission is not an endorsement.

DeepCoder: Learning to Write Programs — mentioned by homarp

Thu, 23 Feb 2017 06:54:02 GMT

Arxiv link https://arxiv.org/abs/1611.01989

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

DeepCoder: Learning to Write Programs — HN submission

Wed, 30 Nov 2016 07:30:21 GMT

DeepCoder: Learning to Write Programs

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