Artificial Intelligence for Games

Similar documents
AI in Computer Games. AI in Computer Games. Goals. Game A(I?) History Game categories

Who am I? AI in Computer Games. Goals. AI in Computer Games. History Game A(I?)

History and Philosophical Underpinnings

Principles of Computer Game Design and Implementation. Lecture 20

Artificial Intelligence

What is Artificial Intelligence? Alternate Definitions (Russell + Norvig) Human intelligence

Building a Better Battle The Halo 3 AI Objectives System

ARTIFICIAL INTELLIGENCE (CS 370D)

Game Artificial Intelligence ( CS 4731/7632 )

TGD3351 Game Algorithms TGP2281 Games Programming III. in my own words, better known as Game AI

TGD3351 Game Algorithms TGP2281 Games Programming III. in my own words, better known as Game AI

Artificial Intelligence Paper Presentation

CSC384 Intro to Artificial Intelligence* *The following slides are based on Fahiem Bacchus course lecture notes.

Outline. Introduction to AI. Artificial Intelligence. What is an AI? What is an AI? Agents Environments

CPS331 Lecture: Agents and Robots last revised November 18, 2016

CS 440 / ECE 448 Introduction to Artificial Intelligence Spring 2010 Lecture #5

Outline. What is AI? A brief history of AI State of the art

RoboCup. Presented by Shane Murphy April 24, 2003

Adjustable Group Behavior of Agents in Action-based Games

CSC 550: Introduction to Artificial Intelligence. Fall 2004

Game Tree Search. CSC384: Introduction to Artificial Intelligence. Generalizing Search Problem. General Games. What makes something a game?

Artificial Intelligence (AI) Artificial Intelligence Part I. Intelligence (wikipedia) AI (wikipedia) ! What is intelligence?

Principles of Computer Game Design and Implementation. Lecture 29

Algorithms for Data Structures: Search for Games. Phillip Smith 27/11/13

CPS331 Lecture: Agents and Robots last revised April 27, 2012

Artificial Intelligence. Minimax and alpha-beta pruning

Playful AI Education. Todd W. Neller Gettysburg College

Conversion Masters in IT (MIT) AI as Representation and Search. (Representation and Search Strategies) Lecture 002. Sandro Spina

Service Robots in an Intelligent House

Random Administrivia. In CMC 306 on Monday for LISP lab

Hierarchical Controller for Robotic Soccer

Computer Science and Information Technology. Spring 2009 Jane Hsu

Philosophy. AI Slides (5e) c Lin

Playing Games. Henry Z. Lo. June 23, We consider writing AI to play games with the following properties:

COS402 Artificial Intelligence Fall, Lecture I: Introduction

CS 480: GAME AI INTRODUCTION TO GAME AI. 4/3/2012 Santiago Ontañón

Extending the STRADA Framework to Design an AI for ORTS

Integrating Learning in a Multi-Scale Agent

FU-Fighters. The Soccer Robots of Freie Universität Berlin. Why RoboCup? What is RoboCup?

UNIT 13A AI: Games & Search Strategies

Outline. Agents and environments Rationality PEAS (Performance measure, Environment, Actuators, Sensors) Environment types Agent types

VI51 Project Subjects

Module 3. Problem Solving using Search- (Two agent) Version 2 CSE IIT, Kharagpur

the gamedesigninitiative at cornell university Lecture 20 Optimizing Behavior

Overview Agents, environments, typical components

What is AI? AI is the reproduction of human reasoning and intelligent behavior by computational methods. an attempt of. Intelligent behavior Computer

CS 1571 Introduction to AI Lecture 12. Adversarial search. CS 1571 Intro to AI. Announcements

FPS Assignment Call of Duty 4

Adversarial Search (Game Playing)

Computer Science Faculty Publications

COMP3211 Project. Artificial Intelligence for Tron game. Group 7. Chiu Ka Wa ( ) Chun Wai Wong ( ) Ku Chun Kit ( )

CSC321 Lecture 23: Go

COMP219: Artificial Intelligence. Lecture 2: AI Problems and Applications

Applying Goal-Driven Autonomy to StarCraft

Creating a Poker Playing Program Using Evolutionary Computation

Artificial Intelligence for Social Impact. February 8, 2018 Dr. Cara LaPointe Senior Fellow Georgetown University

Game Tree Search. Generalizing Search Problems. Two-person Zero-Sum Games. Generalizing Search Problems. CSC384: Intro to Artificial Intelligence

CS 354R: Computer Game Technology

Google DeepMind s AlphaGo vs. world Go champion Lee Sedol

CSC384: Introduction to Artificial Intelligence. Game Tree Search

Today s Topics. Video Game AI: Lecture 2 History of Game AI. Pong (1972) A selective history of video game AI

The Nature of Informatics

the question of whether computers can think is like the question of whether submarines can swim -- Dijkstra

Game-playing: DeepBlue and AlphaGo

AI for Autonomous Ships Challenges in Design and Validation

Capturing and Adapting Traces for Character Control in Computer Role Playing Games

STRATEGO EXPERT SYSTEM SHELL

Non-Deterministic AI in Games. Sai Raghunandan G Master of Science Computer Animation and Visual Effects. November, 2013

Artificial Intelligence

the question of whether computers can think is like the question of whether submarines can swim -- Dijkstra

Discussion of Emergent Strategy

CPS331 Lecture: Intelligent Agents last revised July 25, 2018

ES 492: SCIENCE IN THE MOVIES

CS510 \ Lecture Ariel Stolerman

The Principles Of A.I Alphago

Welcome to CSC384: Intro to Artificial MAN.

Tac 3 Feedback. Movement too sensitive/not sensitive enough Play around with it until you find something smooth

CS 387/680: GAME AI AI FOR FIRST-PERSON SHOOTERS

Your Name and ID. (a) ( 3 points) Breadth First Search is complete even if zero step-costs are allowed.

Coevolution and turnbased games

Interacting Agent Based Systems

Artificial Intelligence: Implications for Autonomous Weapons. Stuart Russell University of California, Berkeley

Grading Delays. We don t have permission to grade you (yet) We re working with tstaff on a solution We ll get grades back to you as soon as we can

mywbut.com Two agent games : alpha beta pruning

CS 380: ARTIFICIAL INTELLIGENCE MONTE CARLO SEARCH. Santiago Ontañón

CS 4700: Foundations of Artificial Intelligence

A CBR/RL system for learning micromanagement in real-time strategy games

COS 402 Machine Learning and Artificial Intelligence Fall Lecture 1: Intro

5.4 Imperfect, Real-Time Decisions

CS 229 Final Project: Using Reinforcement Learning to Play Othello

CMSC 372 Artificial Intelligence. Fall Administrivia

Adversarial Search and Game Theory. CS 510 Lecture 5 October 26, 2017

Strategic and Tactical Reasoning with Waypoints Lars Lidén Valve Software

Elements of Artificial Intelligence and Expert Systems

Behaviour Patterns Evolution on Individual and Group Level. Stanislav Slušný, Roman Neruda, Petra Vidnerová. CIMMACS 07, December 14, Tenerife

Evolutionary Computation for Creativity and Intelligence. By Darwin Johnson, Alice Quintanilla, and Isabel Tweraser

INTRODUCTION TO GAME AI

Artificial Intelligence. 4. Game Playing. Prof. Bojana Dalbelo Bašić Assoc. Prof. Jan Šnajder

Plan for the 2nd hour. What is AI. Acting humanly: The Turing test. EDAF70: Applied Artificial Intelligence Agents (Chapter 2 of AIMA)

AI in Tabletop Games. Team 13 Josh Charnetsky Zachary Koch CSE Professor Anita Wasilewska

Transcription:

Artificial Intelligence for Games

CSC404: Video Game Design Elias Adum Let s talk about AI

Artificial Intelligence AI is the field of creating intelligent behaviour in machines. Intelligence understood to be measured relatives to humans. Labeled as gameplay from a developer s point of view. How do you measure intelligence? Combination of: Perception Processing Expression Turing Test:

CSC404: Video Game Design Elias Adum Areas of AI Perception Language Vision Processing Searching Planning Game Trees Learning Neural Networks

AI Entities The purpose of AI is to create entities that are able to act independently of human interaction In Games, these are usually called non-player characters or NPCs NPCs need to have the following properties: Autonomy: no direct involvement to perform duties Reactivity: perceive and react to environment Proactivity: Goal-directed behaviour Sociability: interaction with other agents

Autonomy Forza Drivatars Paths are determined using waypoints for the NPCs to follow. The NPCs adapt to use those waypoints to: Block the player from passing them Pass the player Forza uses Machine learning to emulate real player behaviour in NPCs. Collect data from human players Apply data it to the course Feed data to physics engine through a virtual controller Currently using big data to analyze drivatars You might encounter friends drivatars in your races

Autonomy Forza Drivatars CSC404: Video Game Design Elias Adum

Reactivity Civilization 5 Each Civ leader has a unique game plan Depends on leader s traits Loyalty, forgiveness, competitiveness, etc Traits ranking: 1 10 AI must adapt to the environment s conditions Produce more units during war Figure out trade routes with allies Naval vs ground exploration Blackboard System Specialists contribute to the engine s decision War AI, Trade AI, Science AI, etc

Reactivity Civilization 5 CSC404: Video Game Design Elias Adum

CSC404: Video Game Design Elias Adum Proactivity PES 2013

CSC404: Video Game Design Elias Adum Sociability The Sims

CSC404: Video Game Design Elias Adum Multi-Agent Applications RoboCup: Robot soccer league By 2050, have a team of autonomous robots beat a team of humans Video Games: Sports games. AI has to coordinate all team members not just single players

CSC404: Video Game Design Elias Adum Types of AI in Games Natural Language Computer Vision Planning Machine Learning

Natural Language - Façade Façade is an interactive story where the player is invited to the apartment of Grace and Trip, a couple who has a relationship breakdown. The player can coach them using full typed sentences. Innovations in Game AI: The player interacts with the game by tying text into a natural language parser. The underlying behavior language provides ways to specify the behavior of characters in terms of a dynamic story. https://aigamedev.com/open/highlights/top-ai-games/#3faade

Natural Language - Façade CSC404: Video Game Design Elias Adum

CSC404: Video Game Design Elias Adum Computer Vision Kinect Other examples?

Planning F.E.A.R. F.E.A.R. First Encounter Assault Recon is a first-person shooter where the player helps contain supernatural phenomenon and, naturally, armies of cloned soldiers. Innovations in Game AI The AI uses a planner to generate context-sensitive behaviors, the first time in a mainstream game. This technology used as a reference for many studios still today. The enemies are capable of using the environment very cleverly, finding cover behind tables, tipping bookshelves, opening doors, crashing through windows Squad tactics are used to great effect. The enemies perform flanking maneuvers, use suppression fire, etc. https://aigamedev.com/open/highlights/top-ai-games/#2fear

Planning F.E.A.R. CSC404: Video Game Design Elias Adum

Planning The Division The Division is an online only open world 3 rd person shooter Innovations in Game AI Player holds down cover button and character will automatically move to new cover spot Game draws a thin white line to indicate exactly how the player will get there Automatic process using AI

Planning The Division CSC404: Video Game Design Elias Adum

Neural Networks Tries to loosely model how the brain neurons are connected to each other Each node is defined by it s input function and an activation function that produces an output The goal of the neural network is to solve problems in the same way that the human brain would Useful when teaching an AI to learn Reinforced learning used in games

Machine Learning Black & White Black & White is a god game which includes elements of artificial life simulations and some strategy. The player rules over an island populated by various tribes, and can teach a large animal-like creature to do his/her bidding. Innovations in Game AI The gameplay is focused on the interaction with a large AI creature which can learn from examples, and takes reward and punishment. The design integrates artificial life within the context of a strategy game. The engine uses a solid AI architecture, rooted in cognitive science. Machine learning techniques such as decision trees and neural networks are used with great success. https://aigamedev.com/open/highlights/top-ai-games/#1blackwhite

Machine Learning Creatures Creatures is an artificial life program where the user hatches small furry animals and teaches them how to behave. These Norms can talk, feed themselves, and protect themselves against vicious creatures. Innovations in Game AI It s the first popular application of machine learning into an interactive simulation. Neural networks are used by the creatures to learn what to do. The game is regarded as a breakthrough in research, which aims to model the behavior of creatures interacting with their environment. https://aigamedev.com/open/highlights/top-ai-games/#5creatures

Machine Learning Mar I/O CSC404: Video Game Design Elias Adum

AI in Mobile Games Plague Inc Plague Inc. is a simulation mobile video game. The aim of the game is to destroy the human race with a Plague that you genetically engineer Different types of Plagues (viruses, bacteria, parasite). Each has their own behaviour Dev modeled the spread of infectious diseases inside the game Stayed at the top of the paid charts for over 2 weeks on both iphone and ipad

AI in Mobile Games Plague Inc

AI in Mobile Games Uniwar HD

Some Useful Algorithms & Data Structures

Minimax Recursive algorithm for choosing the next move in a two-player game Works if algorithm has perfect information Chess Similar to the human thought process of saying, "OK, if I make this move, then my opponent can only make two moves, and each of those would let me win. So this is the right move to make."

A* Informed search algorithm (best-first search). Searches among all possible paths to solution Performance implications? Minimize: f(n) = g(n) + h(n) h(n) problem specific heuristic Widely used in games Path finding and traversal

Finite State Machines CSC404: Video Game Design Elias Adum

Behaviour Trees Similar to FSM with the key difference that the main building block of a BT is a task rather than a state Controls the flow of the decision making of an AI entity Represented as Root node no parent, exactly 1 child Control flow nodes 1 parent, and at least 1 child Execution nodes (tasks) 1 parent, no children On each tick, execution nodes return Running Success Failure

BT - Example CSC404: Video Game Design Elias Adum

BT - Example # 2 CSC404: Video Game Design Elias Adum

BT Example # 3 Alien Isolation Alien Isolation is a horror/survival game based on the movie Alien Xenomorph BT can be outsmarted but, It s hard to predict where and when it might appear It has strong sensing capabilities (hearing) Director system to keep it near the player

BT Example # 3 Alien Isolation