ZeRO & Fastest BERT: Increasing the scale and speed of deep learning training in DeepSpeed webinar
The latest trend in AI is that larger natural language models provide better accuracy; however, larger models are difficult to train because of cost, time, and ease of code integration. With the goal of advancing…
MineRL Competition 2020
Starting July 1st 2020, we are holding a competition on sample-efficient reinforcement learning using human priors as a part of NeurIPS 2020 competition track. In our competition, participants develop a system to obtain a diamond…
eXperiment Tools (XT)
eXperiment Tools (XT) is a command line interface (CLI) and API for running, managing, and analyzing ML experiments. The goal of XT is to enable users to effortlessly organize and scale their ML experiments with…
ONNX Runtime
ONNX Runtime is a cross-platform inferencing and training accelerator compatible with many popular ML/DNN frameworks, including PyTorch, TensorFlow/Keras, scikit-learn, and more. aka.ms/onnxruntime
Graspologic
A graph, or network, provides a mathematically intuitive representation of data with some sort of relationship between items. For example, a social network can be represented as a graph by considering all participants in the…
What Kind of Computation is Human Cognition? A Brief History of Thought (Episode 2/2)
Since the naming of the field in 1956, AI has been dominated first by symbolic rule-based models, then early-generation neural (or “connectionist”) models, then probabilistic models operating over symbolic structures, and now deep learning models.…
ICML 2020 highlights: A Transformer-based RL agent, causal ML for increased privacy, and more
With over 50 papers from Microsoft accepted at this year’s International Conference on Machine Learning (ICML 2020) (opens in new tab), a number of which were presented in virtual workshops, Microsoft researchers are in full…
Situated Interaction in the Open World: New Systems and Challenges
A presentation to the graduate seminar on AI at University of Washington, November 30, 2018. “In this talk, I will introduce a research effort we call “Situated Interaction,” in which we strive to design and…