A significant proportion of our time in this last period has been continued work in MCTS (Monte Carlo Tree Search). Our testbed for this has been the game Spades but our eye has been on finding a good generic formula for driving this for porting to all future card games. This has proved a generally good generic method but has characteristics that need mastering to make it a good portable method, so has taken longer than expected. One article this month looks at what our partners at York have discovered in this domain. The other article takes a sideways look at some lessons learned from our development history.
Article: Reducing the burden of knowledge: Simulation-based methods in imperfect information games: The York group digs deep
Article: Lessons from the Road Championship level AI development: The long road to a winning formula.
The 3rd Party Engine resource: AI Factory has many game engines, all with the same interface for every platform.
Custom AI for 3rd parties: With our background in AI, we are involved in producing customised AI components for 3rd party developers.
Quality 3D products: We already have a range of very high quality 3D products with realistic visual effects.
News round-up: General news about company activities and new products.
AI Factory home: This links through to the main AI Factory website. From there you can review our products and access our engineering documentation.
Future articles scheduled:
Mixing inference and MCTS
Managing a generic turn-based game architecture
Solving Hearts: Analysis of this imperfect information game
We value your feedback. Please click here to reply to this newsletter