Stochastic Mixture-of-Experts
This PyTorch package implements Taming Sparsely Activated Transformer with Stochastic Experts.
Discover an index of datasets, SDKs, APIs and open-source tools developed by Microsoft researchers and shared with the global academic community below. These experimental technologies—available through Azure AI Foundry Labs (opens in new tab)—offer a glimpse into the future of AI innovation.
This PyTorch package implements Taming Sparsely Activated Transformer with Stochastic Experts.
This PyTorch package implements Very Deep Transformers for Neural Machine Translation, to stabilize the large scale language model and neural machine translation training, as described in: Very deep transformers for neural machine translation
This repository contains necessary data associated with the Analyzing the Effects of Reasoning Types on Cross-Lingual Transfer Performance (opens in new tab), published in EMNLP 2021 Multilingual Representation Learning workshop (opens in new tab). This data is…
This is the implementation of the paper Meta Self-training for Few-shot Neural Sequence Labeling. MetaST is short for meta-learning for self-training.
This Python package implements algorithms for online learning under delay using optimistic hints. More details on the algorithms and their regret properties can be found in the manuscript Online Learning with Optimism and Delay.
Many modern AI algorithms are known to be data-hungry, whereas human decision-making is much more efficient. The human can reason under uncertainty, actively acquire valuable information from the world to reduce uncertainty, and make personalized…
The subseasonal_toolkit package provides implementations of the subseasonal forecasting toolkit models, machine learning models, and meteorological baselines presented in the preprint Learned Benchmarks for Subseasonal Forecasting Soukayna Mouatadid, Paulo Orenstein, Genevieve Flaspohler, Miruna Oprescu, Judah…
This repository contains a setup for running a distributed deployment of the Gadgetron in a Kubernetes cluster. The setup has been developed for and tested with Azure Kubernetes Service (AKS) but should work on other…
The HoloLens Navigation for Robots project is an open source repository to be included as a release in the Applied Robotics Research Library. It is a set of sample code to build software capable of…