HackerPapers

arXiv:1605.06640cs.NE

Programming with a Differentiable Forth Interpreter

Matko Bošnjak, Tim Rocktäschel, Jason Naradowsky, Sebastian Riedel

Abstract

Given that in practice training data is scarce for all but a small set of problems, a core question is how to incorporate prior knowledge into a model. In this paper, we consider the case of prior procedural knowledge for neural networks, such as knowing how a program should traverse a sequence, but not what local actions should be performed at each step. To this end, we present an end-to-end differentiable interpreter for the programming language Forth which enables programmers to write program sketches with slots that can be filled with behaviour trained from program input-output data. We can optimise this behaviour directly through gradient descent techniques on user-specified objectives, and also integrate the program into any larger neural computation graph. We show empirically that our interpreter is able to effectively leverage different levels of prior program structure and learn complex behaviours such as sequence sorting and addition. When connected to outputs of an LSTM and trained jointly, our interpreter achieves state-of-the-art accuracy for end-to-end reasoning about quantities expressed in natural language stories.

arXiv comments
34th International Conference on Machine Learning (ICML 2017)
Journal reference
DOI
Subjects
cs.NENeural and Evolutionary Computing cs.AIArtificial Intelligence cs.LGMachine Learning

On Hacker News

Linked by 4 identifiable commenters in 5 comments across 5 threads, from Jul 2016 to Aug 2022. Found in 2 HN submissions from May 2016 to Jan 2018. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments5 threads

  1. Differentiable programming from scratch 1 comment

    • Agreed about the binomial formula, but don't you need induction to prove the binomial theorem? That is, if you have some sort of set that's like the naturals except that Peano's axiom of induction doesn't apply (such as the naturals plus Alberto and Cristina, who are one…

      kragenView on HNParent context

  2. Linear logic and deep learning [pdf] 1 comment

    • What do you think of 'Programming with a Differentiable Forth Interpreter'[0] and 'A Neural Forth Abstract Machine'[1]? I don't know if they are related to your work, but both are on my to-read pile. [0]: https://arxiv.org/abs/1605.06640 [1]…

      twouhmView on HNParent context

  3. How AI based programming could work 1 comment

    • The author proposes a lot of vague ideas in this article (for example "I believe one of the biggest problems is the use of Error Propagation and Gradient Descent") without references or any solid proofs why they are necessary to solve the proposed program (Automate programming…

      sapphireblueView on HNParent context

  4. The future is fewer people writing code? 1 comment

    • There are signs, if you know were to look for them: http://arxiv.org/abs/1510.07211 https://arxiv.org/abs/1410.4615 http://arxiv.org/abs/1605.06640 https://arxiv.org/abs/1410.5401

      sapphireblueView on HNParent context

  5. Teaching an AI to write Python code with Python code 1 comment

    • So it's just a char-RNN? I was hoping this was using one of the seq2seq frameworks for actually writing code, like https://arxiv.org/abs/1605.06640

      aab0View on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.