The empirical success of deep learning has posed significant challenges to machine learning theory: Why can we efficiently train neural networks with gradient descent despite its highly non-convex optimization landscape? Why do over-parametrized…
Imagine being a doctor with a precocious resident permanently by your side, helping you to identify the best treatment path for your patients. That reality may close at hand.
TL;DR: We propose controllable counterfactuals (CoCo) to evaluate dialogue state tracking (DST) models on novel scenarios, which results in significant performance drop of up to 30.8% for state-of-the-art DST models. Using CoCo for…
How do you wrap your arms around a problem as big, complex, and consequential to humanity as climate change? At the center of all the solutions is sustainable technology.