Presentation is loading. Please wait.

Presentation is loading. Please wait.

Machine Learning A Quick look Sources:

Similar presentations


Presentation on theme: "Machine Learning A Quick look Sources:"— Presentation transcript:

1 Machine Learning A Quick look Sources:
Artificial Intelligence – Russell & Norvig Artifical Intelligence - Luger By: Héctor Muñoz-Avila

2 What Is Machine Learning?
“Logic is not the end of wisdom, it is just the beginning” --- Spock time Environment Knowledge Environment System System Action1 Action2 Knowledge changed

3 Learning: The Big Picture
Two forms of learning: Supervised: the input and output of the learning component can be perceived (for example: friendly teacher) Unsupervised: there is no hint about the correct answers of the learning component (for example to find clusters of data)

4 Offline vs. Online Learning
11-Apr-17 Offline vs. Online Learning Online – during gameplay Adapt to player tactics Avoid repetition of mistakes Requirements: computationally cheap, effective, robust, fast learning (Spronck 2004) Offline - before the game is released Devise new tactics Discover exploits Two different types of learning in games: online and offline: Online learning takes place during gameplay against human players. The main applications for online learning are to adapt to player tactics and playing styles and avoid the repetition of mistakes. However, online learning has rarely been applied in commercial games because most game companies are not comfortable with shipping their games with an online learning technique included in the game. What if the NPC learns something stupid? What if the learning takes too long or is too computational expensive? Therefore we set a couple of requirements for online learning: Computationally cheap: Should not disturb flow of gameplay. Effective: Should not generate too many bad inferior opponents. Robust: Should be able to deal with randomness incorporated in games. For instance you don’t want permanently unlearn a specific behavior because it performed badly at some point. The reason for the bad performance could be subscribed to chance. Fast learning: Should lead to results quickly. Don’t want to bore the players with slow learning AI. Offline learning can be applied before a games is being released. Typically, offline learning is used to explore new game tactics or to find exploits.

5 Classification (According to the language representation)
Symbolic Version Space Decision Trees Explanation-Based Learning Sub-symbolic Reinforcement Learning Connectionist Evolutionary

6 Classification (According to the language representation)
Symbolic Version Space Decision Trees Explanation-Based Learning Sub-symbolic Reinforcement Learning Connectionist Evolutionary

7 Version Space Idea: Learn a concept from a group of instances, some positive and some negative Example: target: obj(Size,Color,Shape) Size = {large, small} Color = {red, white, blue} Shape = {ball, brick, cube} Instances: +: obj(large,white,ball) obj(small,blue,ball) −: obj(small,red,brick) obj(large,blue,cube) Two extremes (temptative) solutions: obj(X,Y,Z) obj(large,white,ball) obj(small,blue,ball) … too general too specific concept space obj(large,Y,ball) obj(small,Y,ball) obj(X,Y,ball)

8 How Version Space Works
If we consider only positives If we consider positive and negatives + + + + + + + + + + + + + + + + What is the role of the negative instances? to help prevent over-generalizations

9 Classification (According to the language representation)
Symbolic Version Space Decision Trees Explanation-Based Learning Sub-symbolic Reinforcement Learning Connectionist Evolutionary

10 Explanation-Based learning
C A B C A B C B A B A C ? B A ? C ? B A B C C B A B A C A C A A B C B C C B A A B C Can we avoid making this error again?

11 Explanation-Based learning (2)
C A B ? B A ? C ? A B C C B A B Possible rule: If the initial state is this and the final state is this, don’t do that A C More sensible rule: don’t stack anything above a block, if the block has to be free in the final state

12 Classification (According to the language representation)
Symbolic Version Space Decision Trees Explanation-Based Learning Sub-symbolic Reinforcement Learning Connectionist Evolutionary

13 Motivation # 1: Analysis Tool
Suppose that a company have a data base of sales data, lots of sales data How can that company’s CEO use this data to figure out an effective sales strategy Safeway, Giant, etc cards: what is that for?

14 Motivation # 1: Analysis Tool (cont’d)
Sales data induction Decision Tree Ex’ple Bar Fri Hun Pat Type Res wait x1 no yes some french x4 full thai x5 x6 x7 x8 x9 x10 x11 “if buyer is male & and age between & married then he buys sport magazines”

15 The Knowledge Base in Expert Systems
A knowledge base consists of a collection of IF-THEN rules: if buyer is male & age between & married then he buys sport magazines if buyer is male & age between 18-30 then he buys PC games magazines Knowledge bases of fielded expert systems contain hundreds and sometimes even thousands such rules. Frequently rules are contradictory and/or overlap

16 Main Drawback of Expert Systems: The Knowledge Acquisition Bottle-Neck
The main problem of expert systems is acquiring knowledge from human specialist is a difficult, cumbersome and long activity. Name KB #Rules Const. time (man/years) Maint. time MYCIN KA 500 10 N/A XCON 2500 18 3 KB = Knowledge Base KA = Knowledge Acquisition

17 Motivation # 2: Avoid Knowledge Acquisition Bottle-Neck
GASOIL is an expert system for designing gas/oil separation systems stationed of-shore The design depends on multiple factors including: proportions of gas, oil and water, flow rate, pressure, density, viscosity, temperature and others To build that system by hand would had taken 10 person years It took only 3 person-months by using inductive learning! GASOIL saved BP millions of dollars

18 Motivation # 2 : Avoid Knowledge Acquisition Bottle-Neck
Name KB #Rules Const. time (man/years) Maint. time (man/months) MYCIN KA 500 10 N/A XCON 2500 18 3 GASOIL IDT 2800 1 0.1 BMT KA (IDT) 30000+ 9 (0.3) 2 (0.1) KB = Knowledge Base KA = Knowledge Acquisition IDT = Induced Decision Trees

19 Example of a Decision Tree
Patrons? waitEstimate? no yes 0-10 >60 Full no yes none some Alternate? Reservation? Yes 30-60 no yes Hungry? yes No 10-30 Alternate? Yes no Raining? Fri/Sat? No Yes yes no No no Bar? Yes yes

20 Definition of A Decision Tree
A decision tree is a tree where: The leaves are labeled with classifications (if the classification is “yes” or “no”. The tree is called a boolean tree) The non-leaves nodes are labeled with attributes The arcs out of a node labeled with an attribute A are labeled with the possible values of the attribute A

21 Induction Databases: what are the data that matches this pattern?
Induction: what is the pattern that matches these data? induction Data Ex’ple Bar Fri Hun Pat Type Res wait x1 no yes some french x4 full thai x5 x6 x7 x8 x9 x10 x11 pattern

22 Induction of Decision Trees
Objective: find a concise decision tree that agrees with the examples The guiding principle we are going to use is the Ockham’s razor principle: the most likely hypothesis is the simplest one that is consistent with the examples Problem: finding the smallest decision tree is NP-complete However, with simple heuristics we can find a small decision tree (approximations)

23 Induction of Decision Trees: Algorithm
Initially all examples are in the same group Select the attribute that makes the most difference (i.e., for each of the values of the attribute most of the examples are either positive or negative) Group the examples according to each value for the selected attribute Repeat 1 within each group (recursive call)

24 Example Ex’ple Bar Fri Hun Pat Alt Type wait x1 no yes some French x4
full Thai x5 x6 Italian x7 none Burger x8 x9 x10 x11 No

25 IDT: Example Lets compare two candidate attributes: Patrons and Type. Which is a better attribute? Patrons? none X7(-),x11(-) some X1(+),x3(+),x6(+),x8(+) full X4(+),x12(+), x2(-),x5(-),x9(-),x10(-) Type? french X1(+), x5(-) italian X6(+), x10(-) burger X3(+),x12(+), x7(-),x9(-) X4(+),x12(+) x2(-),x11(-) thai

26 Example of a Decision Tree
Patrons? waitEstimate? no yes 0-10 >60 Full no yes none some Alternate? Reservation? Yes 30-60 no yes Hungry? yes No 10-30 Alternate? Yes no Raining? Fri/Sat? No Yes yes no No no Bar? Yes yes

27 Decision Trees in Gaming
Black & White, developed by Lionhead Studios, and released in 2001 used ID3 Used to predict a player’s reaction to a certain creature’s action In this model, a greater feedback value means the creature should attack

28 Decision Trees in Black & White
Example Attributes Target Allegiance Defense Tribe Feedback D1 Friendly Weak Celtic -1.0 D2 Enemy 0.4 D3 Strong Norse D4 -0.2 D5 Greek D6 Medium 0.2 D7 -0.4 D8 Aztec 0.0 D9

29 Decision Trees in Black & White
Allegiance Friendly Enemy Defense -1.0 Weak Strong Medium 0.4 -0.3 0.1 Note that this decision tree does not even use the tribe attribute

30 Decision Trees in Black & White
Now suppose we don’t want the entire decision tree, but we just want the 2 highest feedback values We can create a Boolean expressions, such as ((Allegiance = Enemy) ^ (Defense = Weak)) v ((Allegiance = Enemy) ^ (Defense = Medium))

31 Classification (According to the language representation)
Symbolic Version Space Decision Trees Explanation-Based Learning Sub-symbolic Reinforcement Learning Connectionist Evolutionary


Download ppt "Machine Learning A Quick look Sources:"

Similar presentations


Ads by Google