Center for Applied Scientific Computing

Machine Intelligence Group

CASC’s Machine Intelligence Group was founded in 2020 to create a home base for technical staff and postdocs conducting fundamental and applied research in machine learning (ML) in support of the Laboratory’s national security missions. Our research portfolio includes a broad set of topics stemming from the growing field of scientific ML, including representation learning, adversarial methods, uncertainty quantification, graph learning, generalizability, interpretability, and model calibration.

Group Lead

Dan Merl: Bayesian inference, causal inference, experimental design, statistical computing, probabilistic programming

Research Staff

Rushil Anirudh: generative modeling, robust ML, unsupervised learning, inverse problems in imaging, high-dimensional geometry, graph-based methods, ML for physical sciences

Peer-Timo Bremer: large-scale data analysis, topological techniques, scientific ML, interpretability, data management

Mark Heimann: graph-based methods, unsupervised learning, dimension reduction, kernel methods

Jayaraman (Jay) Thiagarajan: deep learning, unsupervised learning, generative models, inverse problems, uncertainty quantification, meta learning, transfer learning, graph-based methods, explainable AI, dimension reduction, kernel methods

Bhavya Kailkhura: robust ML, adversarial deep learning, non-convex optimization and decentralized control, robust statistics

Irene Kim: deep learning, unsupervised learning, generative models, self-supervised learning, uncertainty quantification, topological data analysis

Shusen Liu: explainable AI, latent representation interpretation, high-dimensional data visualization, information visualization, visual analytics

Sam Sakla: deep learning, computer vision, self-supervised learning, fine-grained classification, object detection, manifold learning, multi-resolution image/signal processing

Jayram Thathachar: memory-augmented neural networks, multimodal reasoning, optimization, big data algorithms, theoretical computer science

Jize Zhang: reliable and robust deep learning, uncertainty quantification, surrogate modeling, Bayesian inference, scientific ML