Machine Learning and Information Processing Laboratory

ConvDecoder

An un-trained neural network with a potential application in accelerated MRI

Robustness-CS

Measuring the robustness of compressive sensing methods (including deep-learning-based ones) for image reconstruction.

DeepDeWedge

Self-supervised deep learning for denoising and missing wedge reconstruction of cryo-ET tomograms

cs_deep_decoder

transformers_for_imaging

early_stopping_double_descent

Code for reproducing figures and results in the paper ``Early stopping in deep networks: Double descent and how to eliminate it''

overparameterized_convolutional_generators

ttt_for_deep_learning_cs

imaging_MLPs

cinemri

noisy_dna_data_storage

Data recovery from millions of noisy reads

candidate_training

Reproducible version of the experiments in "Leveraging inductive bias of neural networks for learning without explicit human annotations"

sample_complexity_ss_recon

recalibrating_conformal_prediction

QTC: Test-time recalibration of conformal predictors under distribution shift based on unlabeled examples

regularization_based_continual_learning

robust_reconstructors_via_jittering

Scaling_Laws_For_Deep_Learning_Based_Image_Reconstruction

channel_normalization

imagenet_candidate_training

Training on noisy Flickr labels without annotators for better classification performance in ImageNet classification problem.

monotonic_risk_relationships

mri_data_diversity

regularization-wise_double_descent

deep_inverse_book.github.io