Filters








652 Hits in 3.8 sec

Leveraging Grammar and Reinforcement Learning for Neural Program Synthesis [article]

Rudy Bunel, Matthew Hausknecht, Jacob Devlin, Rishabh Singh, Pushmeet Kohli
<span title="2018-05-22">2018</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Recent years have seen proposal of a number of neural approaches for program synthesis, many of which adopt a sequence generation paradigm similar to neural machine translation, in which sequence-to-sequence  ...  To address the first limitation, we perform reinforcement learning on top of a supervised model with an objective that explicitly maximizes the likelihood of generating semantically correct programs.  ...  a small number of samples first for supervised training and again for Reinforcement Learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1805.04276v2">arXiv:1805.04276v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/biqt5yzycnhqvmjz2474xklfha">fatcat:biqt5yzycnhqvmjz2474xklfha</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200826063743/https://arxiv.org/pdf/1805.04276v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/15/32/1532fc492eb8eb55b55a4098ec9d7e31c247f2d1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1805.04276v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Programming by Examples: PL Meets ML [chapter]

Sumit Gulwani, Prateek Jain
<span title="">2017</span> <i title="Springer International Publishing"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/2w3awgokqne6te4nvlofavy5a4" style="color: black;">Lecture Notes in Computer Science</a> </i> &nbsp;
We leverage active-learning techniques based on clustering inputs and synthesizing multiple programs. Each of these PBE components leverage both symbolic reasoning and heuristics.  ...  We make the case for synthesizing these heuristics from training data using appropriate machine learning methods.  ...  in this article related to using ML techniques for search and ranking.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-319-71237-6_1">doi:10.1007/978-3-319-71237-6_1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/nou2fnkpt5elfj3ohaunnfmy7y">fatcat:nou2fnkpt5elfj3ohaunnfmy7y</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20180723152500/https://www.microsoft.com/en-us/research/wp-content/uploads/2017/10/aplas17.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c1/d8/c1d865c894bae4f0dd7d19c2e0a8833fc2b088b0.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-319-71237-6_1"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> springer.com </button> </a>

Code2Inv: A Deep Learning Framework for Program Verification [chapter]

Xujie Si, Aaditya Naik, Hanjun Dai, Mayur Naik, Le Song
<span title="">2020</span> <i title="Springer International Publishing"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/2w3awgokqne6te4nvlofavy5a4" style="color: black;">Lecture Notes in Computer Science</a> </i> &nbsp;
We propose a general end-to-end deep learning framework Code2Inv, which takes a verification task and a proof checker as input, and automatically learns a valid proof for the verification task by interacting  ...  We demonstrate the flexibility of Code2Inv by means of two small-scale yet expressive instances: a loop invariant synthesizer for C programs, and a Constrained Horn Clause (CHC) solver.  ...  We thank the reviewers for insightful comments. We thank Elizabeth Dinella, Pardis Pashakhanloo, and Halley Young for feedback on improving the paper.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-030-53291-8_9">doi:10.1007/978-3-030-53291-8_9</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/zwzve6xymvezrie4rdbv6uwj3u">fatcat:zwzve6xymvezrie4rdbv6uwj3u</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200716053648/https://link.springer.com/content/pdf/10.1007%2F978-3-030-53291-8_9.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/9d/0c/9d0ce199ed4bbfe4e6b4c4be238449cd2ba1a4e1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-030-53291-8_9"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> springer.com </button> </a>

Program Synthesis Using Deduction-Guided Reinforcement Learning [chapter]

Yanju Chen, Chenglong Wang, Osbert Bastani, Isil Dillig, Yu Feng
<span title="">2020</span> <i title="Springer International Publishing"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/2w3awgokqne6te4nvlofavy5a4" style="color: black;">Lecture Notes in Computer Science</a> </i> &nbsp;
In this paper, we present a new program synthesis algorithm based on reinforcement learning.  ...  Specifically, we formulate program synthesis as a reinforcement learning problem and propose a new variant of the policy gradient algorithm that can incorporate feedback from a deduction engine into the  ...  Conclusion and Future Work We presented a new program synthesis algorithm based on reinforcement learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-030-53291-8_30">doi:10.1007/978-3-030-53291-8_30</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/dxb5ws5fqvg6bkmsbba7aln35i">fatcat:dxb5ws5fqvg6bkmsbba7aln35i</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200716053252/https://link.springer.com/content/pdf/10.1007%2F978-3-030-53291-8_30.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/f2/26/f226ea2e949899e939bd15aae2227b9dea7c3066.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1007/978-3-030-53291-8_30"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> springer.com </button> </a>

Closed Loop Neural-Symbolic Learning via Integrating Neural Perception, Grammar Parsing, and Symbolic Reasoning [article]

Qing Li, Siyuan Huang, Yining Hong, Yixin Chen, Ying Nian Wu, Song-Chun Zhu
<span title="2020-07-27">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this paper, we address these issues and close the loop of neural-symbolic learning by (1) introducing the grammar model as a symbolic prior to bridge neural perception and symbolic reasoning, and (2  ...  Prior methods learn the neural-symbolic models using reinforcement learning (RL) approaches, which ignore the error propagation in the symbolic reasoning module and thus converge slowly with sparse rewards  ...  We thank Baoxiong Jia for helpful discussion on the generalized Earley Parser. This work reported herein is supported by ARO W911NF1810296, DARPA XAI N66001-17-2-4029, and ONR MURI N00014-16-1-2007.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.06649v2">arXiv:2006.06649v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/nbr5yyhaqzbkjfdb7z2om2db34">fatcat:nbr5yyhaqzbkjfdb7z2om2db34</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200811214641/https://arxiv.org/pdf/2006.06649v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.06649v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Type-driven Neural Programming by Example [article]

Kiara Grouwstra
<span title="2020-09-17">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We propose a way to incorporate programming types into a neural program synthesis approach for PBE.  ...  As a result of this split, programming types had yet to be used in neural program synthesis techniques.  ...  a synthesis grammar and operator set, program synthesis may make for programs that are potentially more efficient, more understandable, which for the machine learning models produced in supervised learning  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.12613v5">arXiv:2008.12613v5</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ejaoo63f7rfzbetmlxg4w4q63a">fatcat:ejaoo63f7rfzbetmlxg4w4q63a</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200920212711/https://arxiv.org/pdf/2008.12613v5.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.12613v5" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Programmatically Interpretable Reinforcement Learning [article]

Abhinav Verma, Vijayaraghavan Murali, Rishabh Singh, Pushmeet Kohli, Swarat Chaudhuri
<span title="2019-04-10">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We present a reinforcement learning framework, called Programmatically Interpretable Reinforcement Learning (PIRL), that is designed to generate interpretable and verifiable agent policies.  ...  Unlike the popular Deep Reinforcement Learning (DRL) paradigm, which represents policies by neural networks, PIRL represents policies using a high-level, domain-specific programming language.  ...  Acknowledgements This research was partially supported by NSF Award CCF-1162076 and DARPA MUSE Award #FA8750-14-2-0270.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1804.02477v3">arXiv:1804.02477v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/3hydz34zurb4xcr35v6hlkmuvi">fatcat:3hydz34zurb4xcr35v6hlkmuvi</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200914203745/https://arxiv.org/pdf/1804.02477v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/36/9d/369d450fa9f137c271141a8da7c7813a71f27e00.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1804.02477v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Learning to Find Proofs and Theorems by Learning to Refine Search Strategies [article]

Jonathan Laurent, André Platzer
<span title="2022-05-27">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We illustrate our approach on the problem of loop invariant synthesis for imperative programs and using neural networks to refine both the teacher and solver strategies.  ...  We propose a new approach to automated theorem proving and deductive program synthesis where an AlphaZero-style agent is self-training to refine a high-level expert strategy expressed as a nondeterministic  ...  Using reinforcement learning for loop invariant synthesis Reinforcement learning has already been applied to the problem of loop invariant synthesis [27, 28] , albeit in a very different way.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2205.14229v1">arXiv:2205.14229v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/com2bcbtt5fp7deymmjzox6hdu">fatcat:com2bcbtt5fp7deymmjzox6hdu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220605153235/https://arxiv.org/pdf/2205.14229v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/f5/dc/f5dc100525d34e88881e3c5688ecbc1f42c6c83d.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2205.14229v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Learning Differentiable Programs with Admissible Neural Heuristics [article]

Ameesh Shah, Eric Zhan, Jennifer J. Sun, Abhinav Verma, Yisong Yue, Swarat Chaudhuri
<span title="2021-03-28">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our experiments show that the algorithms outperform state-of-the-art methods for program learning, and that they discover programmatic classifiers that yield natural interpretations and achieve competitive  ...  Our key innovation is to view various classes of neural networks as continuous relaxations over the space of programs, which can then be used to complete any partial program.  ...  One direction is to extend the approach to richer DSLs and neural heuristic architectures, for example, those suited to reinforcement learning [37] and generative modeling [29] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2007.12101v5">arXiv:2007.12101v5</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/u2f6zkoclfbexbtbicoczaqfpq">fatcat:u2f6zkoclfbexbtbicoczaqfpq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201014143437/https://arxiv.org/pdf/2007.12101v3.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2007.12101v5" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Neural-guided, Bidirectional Program Search for Abstraction and Reasoning [article]

Simon Alford, Anshula Gandhi, Akshay Rangamani, Andrzej Banburski, Tony Wang, Sylee Dandekar, John Chin, Tomaso Poggio, Peter Chin
<span title="2021-10-26">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
More specifically, we extend existing execution-guided program synthesis approaches with deductive reasoning based on function inverse semantics to enable a neural-guided bidirectional search algorithm  ...  We first apply an existing program synthesis system called DreamCoder to create symbolic abstractions out of tasks solved so far, and show how it enables solving of progressively more challenging ARC tasks  ...  the synthesis task via reinforcement learning.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.11536v2">arXiv:2110.11536v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/wtn27tyxrvcy5pjo7mdw676wd4">fatcat:wtn27tyxrvcy5pjo7mdw676wd4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211028121710/https://arxiv.org/pdf/2110.11536v1.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/ab/9b/ab9b9fded6a58ef839cad363e25791d00c4a763a.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.11536v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

PLANS: Robust Program Learning from Neurally Inferred Specifications [article]

Raphaël Dang-Nhu
<span title="2020-06-05">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We introduce PLANS (Program LeArning from Neurally inferred Specifications), a hybrid model for program synthesis from visual observations that gets the best of both worlds, relying on (i) a neural architecture  ...  Recent years have seen the rise of statistical program learning based on neural models as an alternative to traditional rule-based systems for programming by example.  ...  We did not need to use such approaches as synthesis was reasonably fast in the Karel and ViZDoom benchmarks.Programmatic reinforcement learning The idea of representing reinforcement learning policies  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.03312v1">arXiv:2006.03312v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ttfhnfmwqzdpnm6kcgr2vmuv7e">fatcat:ttfhnfmwqzdpnm6kcgr2vmuv7e</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200827191019/https://arxiv.org/pdf/2006.03312v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/56/83/568361dfdf2534bd08edef34c295142fa5c7fd81.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.03312v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

From explanation to synthesis: Compositional program induction for learning from demonstration [article]

Michael Burke, Svetlin Penkov, Subramanian Ramamoorthy
<span title="2019-02-27">2019</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Using this approach, we are successfully able to induce a program for a visuomotor reaching task involving loops and conditionals from a single demonstration and a neural end-to-end model.  ...  We argue that computer program-like control systems are more interpretable than alternative end-to-end learning approaches, and that hybrid systems inherently allow for better generalisation across task  ...  Program synthesis Given the inferred program and controller primitives, program synthesis becomes trivial.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1902.10657v1">arXiv:1902.10657v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/tqiyn22nbbacjd3gfzkb5atnru">fatcat:tqiyn22nbbacjd3gfzkb5atnru</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200930065026/https://arxiv.org/pdf/1902.10657v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/bc/c0/bcc06e2dd34e8b0171daeb64f98df4fab91fc155.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1902.10657v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Robot Action Selection Learning via Layered Dimension Informed Program Synthesis [article]

Jarrett Holtz, Arjun Guha, Joydeep Biswas
<span title="2020-11-12">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Leveraging these insights, we introduce layered dimension-informed program synthesis (LDIPS) - by reasoning about the physical dimensions of state variables, and dimensional constraints on operators, LDIPS  ...  First, ASPs need to reason about physically meaningful quantities derived from the state of the world, and second, there exists a layered structure for composing these policies.  ...  Acknowledgments This work was partially supported by the National Science Foundation under grants CCF-2102291 and CCF-2006404, and by JPMorgan Chase & Co.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.04133v2">arXiv:2008.04133v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/llmbk5qvr5a77hxdwhcsio7gde">fatcat:llmbk5qvr5a77hxdwhcsio7gde</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201119054859/https://arxiv.org/pdf/2008.04133v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/00/5c/005c3fee1c22c4287b10d1edd6dbbd4ce64d6205.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2008.04133v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

From Explanation to Synthesis: Compositional Program Induction for Learning from Demonstration

Michael Burke, Svetlin Valentinov Penkov, Subramanian Ramamoorthy
<span title="2019-06-22">2019</span> <i title="Robotics: Science and Systems Foundation"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/gjhqqq6dgnaupkvp2ckhefvv6i" style="color: black;">Robotics: Science and Systems XV</a> </i> &nbsp;
Using this approach, we are successfully able to induce a program for a visuomotor reaching task involving loops and conditionals from a single demonstration and a neural endto-end model.  ...  In addition, we are able to discover the program used for a tower building task.  ...  Program synthesis Given the inferred program and controller primitives, program synthesis becomes trivial.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.15607/rss.2019.xv.015">doi:10.15607/rss.2019.xv.015</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/rss/BurkePR19.html">dblp:conf/rss/BurkePR19</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/xi5yuy5mgzgklg5m24dpntqzcu">fatcat:xi5yuy5mgzgklg5m24dpntqzcu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201103201952/https://www.pure.ed.ac.uk/ws/files/87641569/From_explanation_to_synthesis_BURKE_DoA300419_AFV.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/2a/33/2a339b6b6b00d72d8558fce4181ec7ef85cf9d27.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.15607/rss.2019.xv.015"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Neural Attribute Machines for Program Generation [article]

Matthew Amodio, Swarat Chaudhuri, Thomas W. Reps
<span title="2021-10-26">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Trained only on sequences from a known grammar, though, they can still struggle to learn rules and constraints of the grammar.  ...  Recurrent neural networks have achieved remarkable success at generating sequences with complex structures, thanks to advances that include richer embeddings of input and cures for vanishing gradients.  ...  grammars [24] , and recurrent neural networks [6] .  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1705.09231v4">arXiv:1705.09231v4</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/hsuxylsn6jh5dcvmlt3mcjwhle">fatcat:hsuxylsn6jh5dcvmlt3mcjwhle</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211101074151/https://arxiv.org/pdf/1705.09231v4.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/3d/6d/3d6d39af1b262c4e1e14ff5c1533b13ac53fb086.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1705.09231v4" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 652 results