SamsungSAILMontreal

Samsung SAIT AI Lab, Montreal's repositories

Generating and Imputing Tabular Data via Diffusion and Flow XGBoost Models

Language:Python113 7 13

Code for "Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models?" [ICML 2023]

Language:ShellMIT29 4 4

Repository for the PopulAtion Parameter Averaging (PAPA) paper

Language:PythonMIT25 30

Code for "Data-Efficient Structured Pruning via Submodular Optimization" [NeurIPS 2022]

Language:Jupyter NotebookMIT7 1 3

Code for "Difference of Submodular Minimization via DC Programming" [ICML 2023]

Language:MATLABMIT3 20

Hyper-Representations as Generative Models: Sampling Unseen Neural Network Weights [NeurIPS 2022]

Simple Learning to Optimize in PyTorch

Language:PythonMIT2 20

On Evaluation Metrics for Graph Generative Models [ICLR 2022]

Multiset-Equivariant Set Prediction with Approximate Implicit Differentiation [ICLR 2022]

Code for "Fairness in Streaming Submodular Maximization over a Matroid Constraint" [ICML 2023]

000

Code for "LayerMerge: Neural Network Depth Compression through Layer Pruning and Merging" [ICML 2024]

000

LoGAH: Predicting 774-Million-Parameter Transformers using Graph HyperNetworks with 1/100 Parameters.

000