Search Results

Principled and Scalable Exploration Techniques for Reinforcement Learning Theme Game AI Project proposed & supervised by Paulo Rauber To discuss whether this project could become your PhD proposal please email: p.rauber@qmul.ac.uk < Back ​ Principled and Scalable Exploration Techniques for Reinforcement Learning Project proposal abstract: Reinforcement learning has received significant attention due to its success in training agents that play popular games such as Go , Starcraft II , Dota 2 , and others. Inefficient exploration, one of the earliest problems recognized in the field, still limits the success of reinforcement learning approaches that do not require domain knowledge. Although techniques like posterior sampling convincingly solve hard exploration problems in simple domains ( https://searchworks.stanford.edu/view/11891201 ), scalable exploration techniques remain elusive. In this project, you will develop principled and scalable exploration techniques based on reducing model uncertainty ( https://arxiv.org/abs/1609.04436 ). Besides benefiting from games as excellent testbeds, this project has the potential to radically improve automated playtesting. Supervisor: Paulo Rauber Based at:

< Back Discrete versus Ordinal Time-Continuous Believability Assessment Link ​ Author(s) C Pacheco, D Melhart, A Liapis, GN Yannakakis, D Perez-Liebana Abstract ​ More info TBA ​ Link

< Back iGGi Con '22 Our annual conference iGGi Con '22 concluded last Wednesday. With around 150 attendees from academia, industry and organisations, the feedback received so far is overwhelmingly positive! Here are some of the highlights: 05 Sep - Internal Day In line with tradition, the first day of the 3-day conference was arranged as an internal event. After an informal start of the day with a year-group catch up and parallel staff meeting, a range of focus workshops addressed the different stages of "iGGi life" for the respective year groups (e.g., Year 1 Welcome, Thesis Writing, Knowledge Exchange, Post-PhD Career Planning, Academic Writing, etc.). A special Thank You goes out to all the supervisors who attended the iGGi Supervisor Community Session! This was followed by the iGGi Plenary Session - in-person for the first time since September 2019! The discussion centred much around the current cost of living crisis, and in summary: a lot more work still needs to be done to adequately address the situation! The evening and night were spent socialising in town (including optional subsequent partying!). 06 Sep - 1st iGGi Con Day The iGGi Con is a public conference and networking event (registration necessary) aimed at industry, organisations and academics who work in the field of digital games, as well as being an opportunity for the iGGi community to get status-updates on everybody's current research work. Posters - an informal session to discuss and talk about the current work of iGGi PG Researchers (PGRs) and some affiliated researchers who exhibited their posters in the foyer of the Ron Cooke Hub. Keynotes - the wonderful Rachel Kowert and the equally wonderful Tom Cole (who is also an iGGi Alumnus) gave their talks to a very inquisitive audience - a summary of which will become available via our YouTube channel . Talks - six talks by selected iGGi PGRs belonged to the categories "Game Psychology", "Game AI", and "Applied Game Design" - speakers were Elena Petrovskaya , Laura Helsby , Bobby Khaleque , Marko Tot , Prasad Sandbhor , Michael Saiger . Industry Panel - four representative from games industry (Rui Craveirinha, Nina Roussakoff, Tom Keane, Kish Hirani) held a panel discussion on "the perceived difference between academia-performed and industry-performed research" which saw vivid audience engagement. After a busy day, everyone chilled out over drinks. 07 Sep - 2nd iGGi Con Day The morning kicked off with four parallel tracks of workshops, namely a Roundtable on competitive gameplay integrity in Collectible Card Games (Sahar Mirhadi, Guilherme Matos De Faria, Sunny Thaicharoen) Qualitative analysis? In MY Excel? It's more likely than you think (Nathan Hughes) a Tutorial on Regret Minimization for optimal play in imperfect information games (Michael Aichmüller) Accessibility beyond access: Using an APX design approach to tackle real world accessibility issues (Callum Deery, Jozef Kulik) Lunch time coincided with the 1:2:1 Industry Meet Session in which ten fabulous individuals who work in the games industry were available for in-person chat. Participating companies included Earcom, Godolphin Games, Player Research, Safe In Our World, Sony/PlayStation, The Chinese Room, UKIE Talks - six talks by selected iGGi PGRs belonged to the categories "User Experience", "Player Decision and Agency", and "Creative AI" - speakers were Francesca Foffano , George Long , Nick Ballou , Nathan Hughes , Kyle Worrall , Amy Smith . Keynote - last but not least the audience had the pleasure to listen to energetic Sahar Asadi's talk titled "From research to production: the journey of content production automation". The conference concluded with closing remarks by general conference chairs Debbie Maxwell and James Walker. A massive Thank You! also to the PGR Organising Team comprising - Sahar Mirhadi, James Goodman, Prasad Sandbhor, Michael Saiger, Florence Smith Nicholls, Francesca Foffano - you did a fantastic job! See you again next year - this time in London! MARK THE DATE: 13+14 September 2023 ​ Previous 12 Sept 2022 Next

iGGi Partners We are excited to be collaborating with a number of industry partners. IGGI works with industry in some of the following ways: ​ Student Industry Knowledge Transfer - this can take many forms, from what looks like a traditional placement, to a short term consultancy, to an ongoing relationship between the student and their industry partner. Student Sponsorship - for some of our students, their relationship with their industry partner is reinforced by sponsorship from the company. This is an excellent demonstration of the strength of the commitment and the success of the collaborations. In Kind Contributions - IGGI industry partners can contribute by attending and/or featuring in our annual conference, offering their time to give talks and masterclasses for our students, or even taking part in our annual game jam! ​ There are many ways for our industry partners to work with IGGI. If you are interested in becoming involved, please do contact us so we can discuss what might be suitable for you. COMIC Research

< Back " Hunt Takes Hare": Theming Games Through Game-Word Vector Translation Link ​ Author(s) Y Rabii, M Cook Abstract ​ More info TBA ​ Link

< Back A comparison of self-play algorithms under a generalized framework Link ​ Author(s) D Hernandez, K Denamganai, S Devlin, S Samothrakis, JA Walker Abstract ​ More info TBA ​ Link

< Back Decentralised Gambling Overview Link ​ Author(s) OJ Scholten, JA Walker, D Zendle Abstract ​ More info TBA ​ Link

< Back Novel Incentives for Phrase Detectives Link ​ Author(s) M Poesio, J Chamberlain, U Kruschwitz, C Madge Abstract ​ More info TBA ​ Link

iGGi Partners We are excited to be collaborating with a number of industry partners. IGGI works with industry in some of the following ways: ​ Student Industry Knowledge Transfer - this can take many forms, from what looks like a traditional placement, to a short term consultancy, to an ongoing relationship between the student and their industry partner. Student Sponsorship - for some of our students, their relationship with their industry partner is reinforced by sponsorship from the company. This is an excellent demonstration of the strength of the commitment and the success of the collaborations. In Kind Contributions - IGGI industry partners can contribute by attending and/or featuring in our annual conference, offering their time to give talks and masterclasses for our students, or even taking part in our annual game jam! ​ There are many ways for our industry partners to work with IGGI. If you are interested in becoming involved, please do contact us so we can discuss what might be suitable for you. National Science and Media Museum

< Back Speeding up genetic algorithm-based game balancing using fitness predictors Link ​ Author(s) M Morosan, R Poli Abstract ​ More info TBA ​ Link

< Back A local approach to forward model learning: Results on the game of life game Link ​ Author(s) SM Lucas, A Dockhorn, V Volz, C Bamford, RD Gaina, I Bravi, ... Abstract ​ More info TBA ​ Link

< Back A compression method for spectral photon map rendering Link ​ Author(s) G Lai, NJ Christensen Abstract ​ More info TBA ​ Link