Neural networks (NNs) are currently changing the computational paradigm on how to
combine data with mathematical laws in physics and engineering in a profound way …

Intelligence relies on an agent's knowledge of what it does not know. This capability can be
assessed based on the quality of joint predictions of labels across multiple inputs. In …

We present evidence of substantial benefit from efficient exploration in gathering human
feedback to improve large language models. In our experiments, an agent sequentially …

Large-scale linear models are ubiquitous throughout machine learning, with contemporary
application as surrogate models for neural network uncertainty quantification; that is, the …

Predictive distributions quantify uncertainties ignored by point estimates. This paper
introduces The Neural Testbed: an open source benchmark for controlled and principled …

At its core, this thesis aims to enhance the practicality of deep learning by improving the
label and training efficiency of deep learning models. To this end, we investigate data subset …

Machine learning has made tremendous strides in the past decades at producing state-of-
the-art results in safety-critical fields such as self-driving vehicles and medicine. Current …