Level UpROME DEVELOPER CONFERENCE

AIVRomeVideoGameLab

LEVEL UP:

improve
your skills

Let's Play

Meet the Devs - 2019

Piero Savastano

Data scientist (AI)

Piero comes from pure research in bio-inspired robotics, he coded his first neural network from scratch at 19 years old.

 

After working in research institutions (La Sapienza, CNR), he moved to freelance on the topics of data visualization and machine learning, mostly focusing on web applications. Among his clients for consulting and teaching are Sony CSL, Greenpeace Italy, AdEspresso.
He is now collaborating with AIV to apply reinforcement learning to videogames.

Reinforcement learning in games

Machine Learning for video games will become standardized and a commodity in the next years. At the moment, the ecosystem is still immature and fragmented, but it is useful foreseeing some of the possibilities it offers.

Participants will be guided in writing a casual game for browser in Javascript, that includes agents able to learn from their experience and from the intercation with the player. The algorithm used is Q-learning, the simpliest form of reinforcement learning.

TALK (PROGRAMMING)

11/05/2019 from 12:00 to 13:00
Track 1

Sign up on Sched