Automatic Learning of Combat Models for RTS Games

Slides:



Advertisements
Similar presentations
De-anonymizing social networks Arvind Narayanan, Vitaly Shmatikov.
Advertisements

PSMAGE: Balanced Map Generation for StarCraft Alberto Uriarte and Santiago Ontañón Drexel University Philadelphia 1/34 August 11, 2013.
A Survey of Real-Time Strategy Game AI Research and Competition in StarCraft Santiago Ontanon, Gabriel Synnaeve, Alberto Uriarte, Florian Richoux, David.
Artificial Intelligence in Real Time Strategy Games Dan Li.
Us vs. It. Tanks vs. Robot ● Cooperative “Boss Fight” ● Tank players must destroy the Robot before it reaches the city limits. ● Robot is controlled by.
1/38 Game-Tree Search over High-Level Game States in RTS Games Alberto Uriarte and Santiago Ontañón Drexel University Philadelphia October 6, 2014.
Commercial Data Processing Lesson 2: The Data Processing Cycle.
Space Rescue Chad Seippel Cory VanHooser. Story 2050 brand new International Space Station Distress call from ISS about “alien attack” No further communication.
Artificial Intelligence in Game Design Intelligent Decision Making and Decision Trees.
Artificial Intelligence in Game Design Introduction to Learning.
Us vs. It. Tanks vs. Robot ● Cooperative “Boss Fight” ● Tank players must destroy the Robot before it reaches the city limits. ● Robot is controlled by.
RED DEAD REVOLVER Artificial Intelligence Critique By Mitchell C. Dodes CIS 588.
1 Abstract This study presents an analysis of two modified fuzzy ARTMAP neural networks. The modifications are first introduced mathematically. Then, the.
Wish Branches A Review of “Wish Branches: Enabling Adaptive and Aggressive Predicated Execution” Russell Dodd - October 24, 2006.
Selection Sort, Insertion Sort, Bubble, & Shellsort
CS107 Introduction to Computer Science Lecture 7, 8 An Introduction to Algorithms: Efficiency of algorithms.
Towards Modelling Information Security with Key-Challenge Petri Nets Teijo Venäläinen
New Mexico Computer Science For All Introduction to Algorithms Maureen Psaila-Dombrowski.
Kiting in RTS Games Using Influence Maps Alberto Uriarte and Santiago Ontañón Drexel University Philadelphia 1/26 October 9, 2012.
Evolving Multi-modal Behavior in NPCs Jacob Schrum – Risto Miikkulainen –
Active Learning for Class Imbalance Problem
Starcraft Opponent Modeling CSE 391: Intro to AI Luciano Cheng.
Study on Genetic Network Programming (GNP) with Learning and Evolution Hirasawa laboratory, Artificial Intelligence section Information architecture field.
Monday, January 31, 2000 Aiming Wu Department of Computing and Information Sciences, KSU Readings: Chown and Dietterich.
StarCraft Learning Algorithms By Logan Yarnell, Steven Raines, and Dean Antel.
1 A method of successive elimination of spurious arguments for effective solution the search- based modelling tasks Oleksandr Samoilenko, Volodymyr Stepashko.
Artificial Intelligence in Game Design Complex Steering Behaviors and Combining Behaviors.
Dead Run James Kelly, Mike Papo, Josh Lovatt. Basic Details Single Player Top Down Action game.
1/27 High-level Representations for Game-Tree Search in RTS Games Alberto Uriarte and Santiago Ontañón Drexel University Philadelphia October 3, 2014.
Introduction to Loops For Loops. Motivation for Using Loops So far, everything we’ve done in MATLAB, you could probably do by hand: Mathematical operations.
RADHA-KRISHNA BALLA 19 FEBRUARY, 2009 UCT for Tactical Assault Battles in Real-Time Strategy Games.
Artificial Intelligence Research in Video Games By Jacob Schrum
Computer Science 101 Fast Algorithms. What Is Really Fast? n O(log 2 n) O(n) O(n 2 )O(2 n )
Artificial Intelligence in Game Design Lecture 8: Complex Steering Behaviors and Combining Behaviors.
Game AI Matthew Hsieh Meng Tran. Computer Games Many different genres  Action  Role Playing  Adventure  Strategy  Simulation  Sports  Racing Each.
1 Ch. 2: Getting Started. 2 About this lecture Study a few simple algorithms for sorting – Insertion Sort – Selection Sort (Exercise) – Merge Sort Show.
Riza Erdem Jappie Klooster Dirk Meulenbelt EVOLVING MULTI-MODAL BEHAVIOR IN NPC S.
CPSC Why do we need Sorting? 2.Complexities of few sorting algorithms ? 3.2-Way Sort 1.2-way external merge sort 2.Cost associated with external.
Artificial Intelligence in Game Design Lecture 20: Hill Climbing and N-Grams.
Resilient Treats OOF as Knocked Down. Hard to kill Treat Obviously Dead result as a Knock Down instead. Wary Add 1d6 when taking In Sight tests. Bad Shot.
World of Wokcraft The very best in Single pan cooking themed fantasy gaming!
1/23 A Benchmark for StarCraft Intelligent Agents Alberto Uriarte and Santiago Ontañón Drexel University Philadelphia November 15, 2015.
Evolutionary Computing Systems Lab (ECSL), University of Nevada, Reno 1 Authors : Siming Liu, Christopher Ballinger, Sushil Louis
Gene prediction in metagenomic fragments: A large scale machine learning approach Katharina J Hoff, Maike Tech, Thomas Lingner, Rolf Daniel, Burkhard Morgenstern.
Understanding AlphaGo. Go Overview Originated in ancient China 2,500 years ago Two players game Goal - surround more territory than the opponent 19X19.
Automatic Learning of Combat Models for RTS Games
Improving Monte Carlo Tree Search Policies in StarCraft
Deep Reinforcement Learning
Introduction Algorithms Order Analysis of Algorithm
Improving Terrain Analysis and Applications to RTS Game AI
Movement in a full and dynamic environment using a limited influence map Paulo Lafeta Ferreira Artificial Intelligence for Games – CS 580 Professor: Steve.
Ho-Chul Cho, Kyung-Joong Kim, Sung-Bae Cho
Space Invaders inspired game
AlphaGo with Deep RL Alpha GO.
CSC 110 – Fluency in Information Technology Chess
Deep reinforcement learning
Roadmap to Programming work, right, fast KISS
Intra-Domain Routing Jacob Strauss September 14, 2006.
Location Prediction and Spatial Data Mining (S. Shekhar)
Heuristics Definition – a heuristic is an inexact algorithm that is based on intuitive and plausible arguments which are “likely” to lead to reasonable.
Workshop II UU Crowd Simulation Framework
National Dong Hwa University
Predicting Unroll Factors Using Supervised Classification
Ch. 2: Getting Started.
Us vs. It.
Appearance monsters' camp computer's base player's base player's
Us vs. It.
Us vs. It.
Aaron Stokes Game Pitch CIS /17/07
Presentation transcript:

Automatic Learning of Combat Models for RTS Games Alberto Uriarte and Santiago Ontañón Drexel University Philadelphia November 16, 2015

Motivation To use a game-tree search algorithm we need a forward model or “simulator”. In some games (like StarCraft) we don’t have such forward model. The most complex part of a forward model for RTS games is the combat.

The goal Fast and high-level combat simulator. Why fast? To use algorithms like MCTS we need to simulate thousands of combats really quick. Why high-level? Even an “attrition game” (an abstraction of a combat game where units cannot move) is EXPTIME1. So this is already a hard problem. A high-level model reduces branching factor. 1 Furtak and Buro (2010)

Proposed combat simulators Input High-level abstraction of units: number of each unit type by player. Output Surviving units. Sustained DPF model Compute how much time each army needs to destroy the other using the Damage Per Frame of each group. Remove the army that took longer to destroy enemy. Remove casualties from winner army using a target policy. Decreased DPF model Compute how much time to kill one enemy’s unit. Remove the unit killed and reduce HP of survivors. Back to point 1 until one army is destroyed.

Proposed combat simulators Input High-level abstraction of units: number of each unit type by player. Output Surviving units. Sustained DPF model Compute how much time each army needs to destroy the other using the Damage Per Frame of each group. Remove the army that took longer to destroy enemy. Remove casualties from winner army using a target policy. PROs Simpler and Faster Decreased DPF model Compute how much time to kill one enemy’s unit. Remove the unit killed a reduce HP of survivors. Back to point 1 until one army is destroyed. PROs Can be stopped at any time to have a prediction after X fames More accurate predictions

Models parameters Parameters machine learned from replay data. Unit DPF Hardcoded Computed using the weapon damage and the time between shots. Learned When a unit is killed compute the (unit’s HP / time attacking unit) / number of attackers Target Policy Hardcoded Sort unit by kill score (resources cost metric). Learned Used the Borda count method to give points towards a unit type each time we make a choice. Parameters machine learned from replay data.

Results Model accuracy after learning from more than 1,500 combats extracted from replays (more details in the paper) Hardcoded Learned Sustained Model 0.861 0.848 Decreased Model 0.905 0.888 Model accuracy and time compared with a low-level model Accuracy Time (sec) Sustained Model 0.874 0.033 Decreased Model 0.885 0.039 SparCraft (AC) 0.891 1.681 SparCraft (NOK-AV) 0.875 1.358 SparCraft (KC) 0.850 6.873 43 times faster!!! Attack Closest No OverKill Attack Value Kiter Closest