Optimistic Actor Critic avoids the pitfalls of greedy exploration in reinforcement learning
| Kamil Ciosek
One of the core directions of Project Malmo is to develop AI capable of rich interactions. Whether that means learning new skills to apply to challenging problems, understanding complex environments, or knowing when to enlist the help of humans, reinforcement…
In the news | Nature
AI takes on popular Minecraft game in machine-learning contest
To see the divide between the best artificial intelligence and the mental capabilities of a seven-year-old child, look no further than the popular video game Minecraft. A young human can learn how to find a rare diamond in the game…
In the news | InfoQ
Microsoft Releases DialogGPT AI Conversation Model
Microsoft Research's Natural Language Processing Group released the dialogue generative pre-trained transformer (DialoGPT), a pre-trained deep-learning natural language processing (NLP) model for automatic conversation response generation. The model was trained on over 147M dialogues and achieves state-of-the-art results on several benchmarks.
In the news | InformationAge
How can organisations use data effectively, according to corporate VP at Microsoft Azure
The largest enterprises have amassed an incredible amount of data from a variety of sources. But, the problem is identifying that data, accessing it and using that most precious asset effectively.
Icebreaker: New model with novel element-wise information acquisition method reduces cost and data needed to train machine learning models
| Cheng Zhang and Sebastian Tschiatschek
In many real-life scenarios, obtaining information is costly, and getting fully observed data is almost impossible. For example, in the recruiting world, obtaining relevant information (in other words, a feature value) for a company could mean performing time-consuming interviews. The…
Logarithmic mapping allows for low discount factors by creating action gaps similar in size
| Harm van Seijen, Mehdi Fatemi, and Arash Tavakoli
While reinforcement learning (RL) has seen significant successes over the past few years, modern deep RL methods are often criticized for how sensitive they are with respect to their hyper-parameters. One such hyper-parameter is the discount factor, which controls how…
Program synthesis and the art of programming by intent with Dr. Sumit Gulwani
Dr. Sumit Gulwani is a programmer’s programmer. Literally. A Partner Research Manager in the Program Synthesis, or PROSE, group at Microsoft Research, Dr. Gulwani is a leading researcher in program synthesis and the inventor of many intent-understanding, programming-by-example and programming-by-natural…
In the news | Search Engine Journal
Bing is Now Utilizing BERT at a Larger Scale Than Google
Bing revealed today that it has been using BERT in search results before Google, and it’s also being used at a larger scale. Google’s use of BERT in search results is currently affecting 10% of search results in the US,…
In the news | Search Engine Land
Bing says it has been applying BERT since April
Bing has been using BERT to improve the quality of search results since April, Microsoft has stated. The transformer models are now applied to every Bing query globally.