machine learning

Benchmarking egocentric multimodal goal inference for assistive wearable agents

We present a benchmark for egocentric multimodal goal inference for assistive wearable agents. This benchmark evaluates the ability of AI systems to infer user goals from egocentric video, audio, and other sensor modalities in real-world scenarios. …

Accelerating scientific discovery with the common task framework

We propose the common task framework as a mechanism for accelerating scientific discovery through collaborative machine learning. This framework establishes shared datasets, evaluation metrics, and benchmarks that enable the AI and scientific …

DigiData: Training and evaluating general-purpose mobile control agents

We present DigiData, a comprehensive framework for training and evaluating general-purpose mobile control agents. This work addresses the challenge of creating AI agents that can navigate and interact with mobile user interfaces to perform tasks …

Challenges in AI for Science

Panel discussion at ICLR 2024 Workshop on AI4DifferentialEquations in Science alongside Max Welling and Shirley Ho

Accelerating look-ahead in Bayesian optimization: Multilevel Monte Carlo is all you need

We propose a multilevel Monte Carlo approach for accelerating look-ahead strategies in Bayesian optimization. Look-ahead acquisition functions, which consider multiple future evaluations, can significantly improve optimization performance but are …

Model reduction for the material point method via an implicit neural representation of the deformation map

This work proposes a model-reduction approach for the material point method on nonlinear manifolds. To represent the low-dimensional nonlinear manifold, we consider an implicit neural representation (INR) that parameterizes the material-point …

Deep Conservation: A latent dynamics model for exact satisfaction of physical conservation laws

This work proposes an approach for latent dynamics learning that exactly enforces physical conservation laws. The method comprises two steps. First, we compute a low-dimensional embedding of the high-dimensional dynamical-system state using deep …

Time-series machine-learning error models for approximate solutions to parameterized dynamical systems

This work proposes a machine-learning framework for modeling the error incurred by approximate solutions to parameterized dynamical systems. In particular, we extend the machine-learning error models (MLEM) framework proposed in [Freno, Carlberg, …

Recovering missing CFD data for high-order discretizations using deep neural networks and dynamics learning

Data I/O poses a significant bottleneck in large-scale CFD simulations; thus, practitioners would like to significantly reduce the number of times the solution is saved to disk, yet retain the ability to recover any field quantity (at any time …

Machine-learning error models for approximate solutions to parameterized systems of nonlinear equations

This work proposes a machine-learning framework for constructing statistical models of errors incurred by approximate solutions to parameterized systems of nonlinear equations. These approximate solutions may arise from early termination of an …