lunduniversity.lu.se

Digit@LTH

Faculty of Engineering, LTH

Denna sida på svenska This page in English

Digit@LTH: Events

CS MSc Thesis Zoom Presentation 20 December 2021

Föreläsning

From: 2021-12-20 09:15 to 10:00
Place: E:2405 (Glasburen) and online via: https://lu-se.zoom.us/j/68249486247
Contact: birger [dot] swahn [at] cs [dot] lth [dot] se
Save event to your calendar


One Computer Science MSc thesis to be presented on 20 December both live and via Zoom

Monday, 20 December there will be a master thesis presentation in Computer Science at Lund University, Faculty of Engineering.

The presentation will take place in E:2405 (Glasburen) and via Zoom at: https://lu-se.zoom.us/j/68249486247

Note to potential opponents: Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation (firstname.lastname@cs.lth.se). Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions are found on this page.


09:15-10:00

Presenters: Daniel Karlsson, Jascha Thiel
Title: Evaluating Curiosity Driven Exploration in a Large Action Space using Starcraft 2
Examiner: Elin Anna Topp
Supervisor: Volker Krueger (LTH)

Starcraft 2 is a PC game that has been addressed more and more within the research of artificial intelligence (AI ) and reinforcement learning. One field of special interest is then the concept of Intrinsic Curiosity Exploration, a reward generated by an Intrinsic Curiosity Model, that incentives exploration called Intrinsic Curiosity. This model is defined as the agent error in predicting the outcome of its action, based on some feature space learned by an inverse dynamics model. Starcraft 2 presents an interesting problem since the action space is large, and many actions have similar outcome. We implement in our thesis an Intrinsic Curiosity Module, with a custom loss function, to work with a well-established reinforcement learning agent, the Deep Q-network. The aim of our research was to learn and be able to play Starcraft 2 in an improved manner.

Link to Zoom presentation: https://lu-se.zoom.us/j/68249486247

Link to popular science summary: TBU