Psychological Sciences Events
[PAST EVENT] David Nader, Computer Science - Dissertation Proposal
Abstract:
Neural Code Models (NCMs) are rapidly progressing from research prototypes to commercial developer tools. As such, understanding the capabilities and limitations of such models is becoming critical. However, the abilities of these models are typically measured using automated metrics that often only reveal a portion of their real-world performance. While, in general, the performance of NCMs appears promising, currently much is unknown about how such models arrive at decisions. To this end, this paper introduces doCode, a post hoc interpretability method specific to NCMs that is capable of explaining model predictions. doCode is based upon causal inference to enable programming language-oriented explanations. While the theoretical underpinnings of doCode are extensible to exploring different model properties, we provide a concrete instantiation that aims to mitigate the impact of spurious correlations by grounding explanations of model behavior in properties of programming languages. To demonstrate the practical benefit of doCode, we illustrate the insights that our framework can provide by performing a case study on two popular deep learning architectures and ten NCMs. The results of this case study illustrate that our studied NCMs are sensitive to changes in code syntax. All our NCMs, except for the BERT-like model, statistically learn to predict tokens related to blocks of code (e.g. brackets, parenthesis, semicolon) with less confounding bias as compared to other programming language constructs. These insights demonstrate the potential of doCode as a useful method to detect and facilitate the elimination of confounding bias in NCMs.
Bio: David N. Palacio is a Ph.D. Candidate in Computer Science at William & Mary, where he is a member of the SEMERU Research Group supervised by Dr. Denys Poshyvanyk. He received his MSc. in Computer Engineering at Universidad Nacional de Colombia (UNAL), Colombia, 2017. His research is concentrated on interpretable methods for deep learning code generators, specifically, on using causal inference to explain deep software models. His fields of interest lie in complexity science, neuroevolution, causal inference, and interpretable machine learning for the study and automation of software engineer processes. More information is available at https://danaderp.github.io/danaderp/ .
Sponsored by: Computer Science