Ian Goodfellow

Adversarial Attacks on Neural Network Policies

2017 · arXiv (Cornell University)

Machine learning classifiers are known to be vulnerable to inputs maliciously constructed by adversaries to force misclassification. Such adversarial examples have been extensively studied in the context of computer vision applications. In this work, we …

MaskGAN: Better Text Generation via Filling in the______

2018 · arXiv (Cornell University)

Neural text generation models are often autoregressive language models or seq2seq models. These models generate text by sampling words sequentially, with each word conditioned on the previous word, and are state-of-the-art for several machine translation …