SPOC lab's repositories
double_descent_uncertainty
Companion repository for the paper "On double-descent in uncertainty quantification in overparametrized models"
gibbs-sampler-neural-networks
Gibbs sampler for neural network posteriors
rectangular_RIE
Code for reproducing the results in "Optimal denoising of rotationally invariant rectangular matrices"
Rigorous-dynamical-mean-field-theory
Code for "Rigorous dynamical mean field theory for stochastic gradient descent methods"
expectation-consistency
Code to compare variants of temperature scaling
FundamentalLimitsMultiIndex
Code repository for the paper: "Fundamental limits of weak learnability in high-dimensional multi-index models"
OneStepGD_asymptotics
code for Asymptotics of feature learning in two-layer networks after one gradient-step
BootstrapAsymptotics
Code for the paper "Analysis of Bootstrap and Subsampling in High-dimensional Regularized Regression"
Diffusion_GMM
Repo for the paper : Analysis of Learning a Flow-based Generative Model from Limited Sample Complexity
Maximum-Average-Submatrix
Code for the RS and 1-RSB solution of the maximum-average submatrix problem
backtracking-dynamical-cavity
A version cavity method that tracks the dynamics of a model backwards from the attractors into their basin.
DAE
Repository for the paper "High Dimensional Asymptotics of Denoising Autoencoders"
DMFT-integration
Code to integrate the DMFT equations
high-dimensional-uncertainty
Companion code for the paper "Theoretical characterization of uncertainty in high-dimensional linear classification"
StochasticBlockModelVariants.jl
Simulation and inference for contextual and neural SBMs