Explore-AI CPT Meetup: NLP for Africa and The Inner Workings of AlphaZero

Welcome to our First Meetup of 2020!

Agenda

17:30 Welcome and Snacks
18:00 Sebastian Bodenstein - AlphaZero/MuZero
18:30 Laura Martinus - Masakhane: NLP for Africa

Speakers:

Sebastian Bodenstein
Talk Title: AlphaZero/MuZero

Abstract:
AlphaZero is the successor to the famous AlphaGo algorithm that beat a champion Go player. Unlike AlphaGo, it requires no human games to initialize its training, and can learn to play other games, such as Chess. This talk will break down how AlphaZero works without assuming any prior reinforcement learning knowledge. It will also look at a recent extension, MuZero, that lifts one of the main limitations of AlphaZero: the need to have an explicit model of the environment. MuZero has achieved stunning results on various reinforcement learning domains, such as learning to play Atari games.

Bio:
Sebastian Bodenstein moved into machine learning after finishing his PhD in Theoretical Physics in 2013. At Wolfram Research, he was one of the two main developers of the Mathematica Neural Net Framework, and the lead developer of the Wolfram Neural Net repository. Recently, he implemented AlphaZero for the DeepMind OpenSpiel reinforcement learning package.

Laura Martinus
Talk Title: 'Masakhane: NLP for Africa'

Abstract:
Africa has over 2000 languages. Despite this, African languages account for a small portion of available resources and NLP publications. This is due to multiple factors, including: a lack of focus from government and funding, discoverability, a lack of community, sheer language complexity, difficulty in reproducing papers and no benchmarks to compare techniques. To begin to address the identified problems, Masakhane, an open-source, continent-wide, distributed, online research effort for machine translation for African languages, was founded. In this talk, we'll discuss our methodology for building the community and spurring research from the African continent, as well as outline the success of the community in terms of addressing the identified problems affecting African NLP.

Bio:
Laura spends her time training future data scientists at Explore. She is particularly interested in using data science to improve living conditions in Africa in terms of access to healthcare and education. Her research interests are in practical applications of NLP, with a focus on creating translation systems for the African languages. When she’s not at work, you can find Laura hiking, doing yoga, or writing NeurIPS papers.