Training and evaluating NBM and SPAM for interpretable machine learning.
Implementation for the paper "Reliable Visual Question Answering Abstain Rather Than Answer Incorrectly" (ECCV 2022: https//arxiv.org/abs/2204.13631).
Holotorch is an optimization framework for differentiable wave-propagation written in PyTorch
This is an official pytorch implementation of Learning To Recognize Procedural Activities with Distant Supervision. In this repository, we provide PyTorch code for training and testing as described in the paper. The proposed distant supervision framework achieves strong generalization performance on step classification, recognition of procedural activities, step forecasting (New Task!) and egocentric activity classification.
Code for paper Learning Audio-Visual Dereverberation
A suite for blackbox variational inference using families of posterior approximations parameterised via learnable weighted summarising (pseudo)data.
Multi-Objective Bayesian Optimization over High-Dimensional Search Spaces
Code for Point-Level Regin Contrast (https//arxiv.org/abs/2202.04639)
PyTorch code for CVPR 2022 paper Unbiased Teacher v2 Semi-supervised Object Detection for Anchor-free and Anchor-based Detectors
A data-driven, fast driving simulator for multi-agent coordination under partial observability.
Code for the paper "TAVA Template-free Animatable Volumetric Actors".
[ICML 2022] "DepthShrinker: A New Compression Paradigm Towards Boosting Real-Hardware Efficiency of Compact Neural Networks", by Yonggan Fu, Haichuan Yang, Jiayi Yuan, Meng Li, Cheng Wan, Raghuraman Krishnamoorthi, Vikas Chandra, Yingyan Lin
Open source code for paper "Denoised MDPs: Learning World Models Better Than the World Itself"
Implementations of robust Dual Curriculum Design (DCD) algorithms for unsupervised environment design.
This repository provides source code, trained neural network model and dataset for our NeuralPassthrough work that is published at SIGGRAPH 2022.
Code for the paper "The Neural Race Reduction Dynamics of Abstraction in Gated Networks"
