source{d} paper reading club
Discuss papers at the intersection of Software Engineering, Programming Languages and Machine Learning communities related to applications of Machine Learning to Code.
Goal
- promote reading papers
- rise awareness on current research directions in PL, SW, ML communities
- create opportunity for collaborations
How it works
Every 2 weeks we pick and discuss a paper.
Next paper is chosen from a list of candidates established at the end of the session.
Anyone can comment on a paper's PDF on GDrive with questions or things that are worth clarifying.
When
Every 2 weeks on Fridays at 4pm CET
Where
- on-line on zoom, ID of the meeting is
974-346-848. - in source{d} office in Madrid
How to participate
- propose/vote for a paper
- join the session online, or in office space
How to organize
A description of the current organization workflow is maintained in
organization-workflow.md.
Past papers
-
2019.11.29 CodeSearchNet Challenge: Evaluating the State of Semantic Code Search. (notes)
-
2019.11.15 When Deep Learning Met Code Search. (notes)
-
2019.10.18 Assessing the Generalizability of code2vec Token Embeddings. (notes)
-
2019.10.04 The Software Heritage Graph Dataset: Public Software Development Under One Roof. Antoine Pietri, first author of the paper joined us. (notes)
-
2019.09.20 End-to-end Deep Learning of Optimization Heuristics. Chris Cummins, author of the paper joined us. (notes)
-
2019.09.06 Topology Adaptive Graph Convolutional Networks. (notes)
-
2019.08.09 Attention Is All You Need. (notes)
-
2019.07.26 Aroma: Code Recommendation via Structural Code Search. (notes)
-
2019.07.12 XLNet: Generalized Autoregressive Pretraining for Language Understanding. (notes)
-
2019.06.28 Import2vec Learning Embeddings for Software Libraries. (notes)
-
2019.06.21 Cross-language clone detection by learning over abstract syntax trees. Daniel Perez, co-author of the paper joined us! (notes)
-
2019.06.14 Coloring Big Graphs with AlphaGoZero. (notes)
-
2019.05.31 Neural Networks for Modeling Source Code Edits and Learning to Represent Edits. (notes)
-
2019.05.17 Maybe Deep Neural Networks are the Best Choice for Modeling Source Code. (notes)
-
2019.05.03 skipped, due to long holidays (moved to the next slot).
-
2019.04.19 A Comprehensive Survey on Graph Neural Networks. (notes)
-
2019.04.05 How Powerful are Graph Neural Networks?. (notes)
-
2019.03.22 Generative Code Modeling with Graphs. (notes)
-
2019.02.22 The Adverse Effects of Code Duplication in Machine Learning Models of Code. (notes)
-
2019.02.08 Structured Neural Summarization. (notes)
-
2019.01.25 Sorting and Transforming Program Repair Ingredients via Deep Learning Code Similarities. Martin Monperrus and Matias Martinez, co-authors of the paper joined us! (notes)
-
2019.01.11 A general reinforcement learning algorithm that masters chess, shogi and Go through self-play. (notes)
-
2018.12.14 Improving Automatic Source Code Summarization via Deep Reinforcement Learning. (notes)
-
2018.11.30 Mining Change Histories for Unknown Systematic Edits. Tim Molderez, first author of the paper, joined us for this session! (notes)
-
2018.11.16 Deep Learning Type Inference. This time Earl T. Barr joined, one of the authors of the paper! (notes, meetup)
-
2018.11.02 Learning to Represent Edits. (notes)
-
2018.10.19 Relational inductive biases, deep learning, and graph networks. (notes)
-
2018.10.5: extra session, Code Vectors: Understanding Programs Through Embedded Abstracted Symbolic Traces. This time Jordan Henkel joined, one of the authors of the paper! (notes, slides)
-
2018.09.28: code2seq: Generating Sequences from Structured Representations of Code by Uri Alon, Eran Yahav and Omer Levy. (notes)
-
2018.09.14: Learning to Represent Programs with Graphs by Miltiadis Allamanis, Marc Brockschmidt and Mahmoud Khademi. (notes)
-
2018.08.31: Intelligent Code Reviews Using Deep Learning by Anshul Gupta and Neel Sundaresan. (notes)
All the past papers we studied are available in the reading club's GDrive.