Computational Mathematics: Publications
Read the latest papers published by our researchers.
Results
-
MoMo: Momentum Models for Adaptive Learning Rates
-
Amortized Variational Inference: When and Why?
-
Heavy-Tailed Class Imbalance and Why Adam Outperforms Gradient Descent on Language Models
-
Delayed rejection Hamiltonian Monte Carlo for sampling multiscale distributions
-
posteriordb: Testing, Benchmarking and Developing Bayesian Inference Algorithms
-
Good Rates From Bad Coordinates: The Exponential Average Time-dependent Rate Approach
-
AstroCLIP: a cross-modal foundation model for galaxies
-
Learning sum of diverse features: computational hardness and efficient gradient-based training for ridge combinations
-
Variational Inference for Uncertainty Quantification: an Analysis of Trade-offs
-
How Truncating Weights Improves Reasoning in Language Models
-
Offline supervised learning vs online direct policy optimization: A comparative study and a unified training paradigm for neural network-based optimal feedback control
-
Contextual Counting: A Mechanistic Study of Transformers on a Quantitative Task
-
Crowdsourcing with Difficulty: A Bayesian Rating Model for Heterogeneous Items
-
Neurosift: DANDI exploration and NWB visualization in the browser
-
Why is parameter averaging beneficial in SGD? An objective smoothing perspective
-
Simulation-Based Stacking
-
GIST: Gibbs self-tuning for locally adaptive Hamiltonian Monte Carlo
-
Improved statistical and computational complexity of the mean-field Langevin dynamics under structured data
-
Efficient convergent boundary integral methods for slender bodies
-
Galaxy clustering analysis with SimBIG and the wavelet scattering transform
-
Escaping saddle points efficiently with occupation-time-adapted perturbation
-
Statistical Component Separation for Targeted Signal Recovery in Noisy Mixtures
-
Nonlinear spiked covariance matrices and signal propagation in deep neural networks
-
For how many iterations should we run Markov chain Monte Carlo?
-
Design of Perfectly Conducting Objects That Are Invisible to an Incident Plane Wave
-
Deep reinforcement learning in finite-horizon to explore the most probable transition pathway
-
Envelopes of Horospheres and Weingarten Surfaces in Hyperbolic 3-Space
-
The magnetic gradient scale length explains why certain plasmas require close external magnetic coils
-
Should Under-parameterized Student Networks Copy or Average Teacher Weights?
-
Generalization in diffusion models arises from geometry-adaptive harmonic representations
-
Scaling Laws for Associative Memories
-
Solving the Scattering Problem for Open Wave-Guides, III: Radiation Conditions and Uniqueness
-
Stochastic Optimal Control Matching
-
Learning High-Dimensional McKean–Vlasov Forward-Backward Stochastic Differential Equations with General Distribution Dependence
-
Directional Smoothness and Gradient Methods: Convergence and Adaptivity
-
Uniform approximation of common Gaussian process kernels using equispaced Fourier grids
-
A new provably stable weighted state redistribution algorithm
-
A new version of the adaptive fast Gauss transform for discrete and continuous sources
-
Implicit Adaptive Mesh Refinement for Dispersive Tsunami Propagation
-
A high-order fast direct solver for surface PDEs
-
Explainable Equivariant Neural Networks for Particle Physics: PELICAN
-
Provable convergence guarantees for black-box variational inference
-
Sharp error estimates for target measure diffusion maps with applications to the committor problem
-
Improving Convergence and Generalization Using Parameter Symmetries
-
Microscopic Theory, Analysis, and Interpretation of Conductance Histograms in Molecular Junctions
-
Bayesian spatial modelling of localised SARS-CoV-2 transmission through mobility networks across England
-
Birth of a Transformer: A Memory Viewpoint
-
On Learning Gaussian Multi-index Models with Gradient Flow