ICT
Video games have come on leaps and bounds in the past decades. The introduction of high levels of complexity, storytelling to rival the best Hollywood films, and online cooperation means that games are gripping players like never before. However, Noor Shaker of the Aalborg University believes that to truly engage people, games should be able to react to the player’s behaviour on the fly to improvise content that both challenges and entertains them
PlayGALe - Playful games with active learning
The rise of video games since the release of Pong in 1972 has been meteoric. With revenues recently topping $100m annually, the industry has now surpassed both films and music financially and is continuing to grow. With video games now a part of people’s lives more than ever
innovative
before, the demand and
entertaining
for more games
is
pushing the medium into territories that gamers thirty years ago could barely imagine. Massively multiplayer online games in which people collaborate online towards shared goals are now commonplace, and VR technology such as the Oculus Rift is edging ever closer to the mainstream. Despite these developments and their increasing complexity, video games
still
tend to be somewhat rigid affairs that cannot diverge from the content that is
48
“Games will be more like playmates in that they can innovate, be creative and react to the player’s actions”
programmed into them. Noor Shaker of the Aalborg University is looking to change this by developing techniques for building new types of playful games in which the system continuously learns from in-game interaction and responds by generating novel content. “The idea is to build games that have more meaningful interactions with the player,” she explains. “Games will be more like playmates in that they can innovate, be creative and react to the player’s actions.” Shaker is using machine-learning techniques based on the concepts of active
learning and artificial curiosity that autonomously choose how to explore the player’s
responses in interactive games.
These will allow games to create new content such as rules, items and levels. For instance, active learning approaches
allow content to be personalised to the behaviour of a specific player. Shaker and her colleagues have been carrying out preliminary investigations experience models that
learning with as few data points as possible. This will make it easier for them to be integrated into the game generation loop,
Insight Publishers | Projects
into building facilitate active
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64