Abstract

Speaker 1: Miltos Allamanis (MSR Cambridge). Title: Learning to Detect Bugs in Source Code with Machine Learning. Abstract: Deep neural networks are succeeding at a range of natural language tasks such as machine translation and text summarization. Recently, the interdisciplinary field of “big code” promises a new set of learnable statistical static analyses. While machine learning tasks on source code have been considered recently, most work in this area does not attempt to capitalize on the unique opportunities offered by its known syntax and structure. In this talk, I discuss how graph neural networks can learn from code’s syntactic and semantic structure to detect variable misuse bugs in code without any external information (e.g. unit tests). (Joint work with Marc Brockschmidt and others).

Speaker 2: Niki Kilbertus (Cambridge university). Title: Learning Independent Causal Mechanisms. Abstract: Independent causal mechanisms are a central concept in the study of causality with implications for machine learning tasks. In this work we develop an algorithm to recover a set of (inverse) independent mechanisms relating a distribution transformed by the mechanisms to a reference distribution. The approach is fully unsupervised and based on a set of experts that compete for data to specialize and extract the mechanisms. We test and analyze the proposed method on a series of experiments based on image transformations. Each expert successfully maps a subset of the transformed data to the original domain, and the learned mechanisms generalize to other domains. We discuss implications for domain transfer and links to recent trends in generative modeling. (Joint work with Giambattista Parascandolo, Mateo Rojas-Carulla, and Bernhard Schölkopf).