Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples · HackerPapers discussions

Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples — mentioned by Cybiote

Wed, 07 Oct 2020 18:57:47 GMT

> What is this sentence supposed to convey? There are examples where it is able to recognize and continue patterns in strings which if manually generated, would have required a FSM. In fact, some of the more impressive examples would require a stack of some sort so I thought I was rather underselling its capabilities in that arena. > as GPT isn't a "state machine" as the latent space is continuous and not finite. Technically speaking, that is impossible since these models leverage floating point numbers and are limited in memory to whatever hidden and self-attention layers. Practically…

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

Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples — mentioned by long

Sat, 12 May 2018 06:05:00 GMT

I think you will like this paper then: https://arxiv.org/abs/1711.09576

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

Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamples — HN submission

Tue, 28 Nov 2017 14:57:36 GMT

Extracting Automata from RNNs Using Queries and Counterexamples [pdf]

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