When More is Less: Incorporating Additional Datasets Can Hurt Performance By Introducing Spurious Correlations

1 code implementation • 8 Aug 2023 • Rhys Compton, Lily Zhang, Aahlad Puli, Rajesh Ranganath

In machine learning, incorporating more data is often seen as a reliable strategy for improving model performance; this work challenges that notion by demonstrating that the addition of external datasets in many cases can hurt the resulting model's performance.

MEDCOD: A Medically-Accurate, Emotive, Diverse, and Controllable Dialog System

1 code implementation • 17 Nov 2021 • Rhys Compton, Ilya Valmianski, Li Deng, Costa Huang, Namit Katariya, Xavier Amatriain, Anitha Kannan

We present MEDCOD, a Medically-Accurate, Emotive, Diverse, and Controllable Dialog system with a unique approach to the natural language generator module.

Sentence

Embedding Java Classes with code2vec: Improvements from Variable Obfuscation

1 code implementation • 6 Apr 2020 • Rhys Compton, Eibe Frank, Panos Patros, Abigail Koay

code2vec is a recently released embedding approach that uses the proxy task of method name prediction to map Java methods to feature vectors.

Code Classification Method name prediction