Contents -- List of Contributors -- Chapter 1 The Iterated Prisoner's Dilemma: 20 Years On Siang Yew Chong, Jan Humble, Graham Kendall, Jiawei Li and Xin Yao -- 1.1. Introduction -- 1.2. Iterated Prisoner's Dilemma -- 1.3. Contents of the Book -- 1.4. Celebrating the 20th Anniversary: The Competitions -- 1.5. Competition Results -- 1.6. Acknowledgements -- Appendix: Software Framework -- References -- Chapter 2 Iterated Prisoner's Dilemma and Evolutinary Game Theory Siang Yew Chong, Jan Humble, Graham Kendall, Jiawei Li and Xin Yao -- 2.1. Introduction -- 2.2. Strategies in IPD Tournaments -- 2.2.1. Heterogeneous TFTs -- 2.2.2. Pavlov (Win-Stay Lose-Shift) -- 2.2.3. Gradual -- 2.2.4. Adaptive strategies -- 2.2.5. Group strategies -- 2.3. Evolutionary Dynamics in Games -- 2.3.1. Evolutionary stable strategy -- 2.3.2. Genetic algorithm -- 2.3.3. Strategies -- 2.3.4. Population -- 2.3.5. Selection scheme -- 2.4. Evolution of Cooperation -- References -- Chapter 3 Learning IPD Strategies Through Co-evolution Siang Yew Chong, Jan Humble, Graham Kendall, Jiawei Li and Xin Yao -- 3.1. Introduction -- 3.2. Co-evolving Strategies for the IPD Game -- 3.2.1. Co-evolutionary Learning Framework -- 3.2.2. Shadow of the Future -- 3.2.3. Issues for Co-evolutionary Learning of IPD Strate- gies -- 3.3. Extending the IPD Game -- 3.3.1. Extending the IPD with More Choices -- 3.3.2. IPD with Noise -- 3.3.3. N-Player IPD -- 3.3.4. Other Extensions -- 3.4. Conclusion and Future Directions -- References -- Chapter 4 How to Design a Strategy to Win an IPD Tournament Jiawei Li -- 4.1. Introduction -- 4.2. Analysis of strategies involved in IPD games -- 4.3. Estimation of possible strategies in an IPD tournament -- 4.4. Interaction with a strategy optimally -- 4.5. Escape from the trap of defection -- 4.6. Adaptive Pavlov and Competition 4 of 2005 IPD tour- nament.

4.7. Discussion and conclusion -- References -- Chapter 5 An Immune Adaptive Agent for the Iterated Prisoner's Dilemma Oscar Alonso and Fernando Ni~no -- 5.1. Introduction -- 5.2. Immune network fundamentals -- 5.3. A general adaptive agent model -- 5.4. Immune agent model -- 5.4.1. Strategy representation -- 5.4.2. Memory -- 5.4.3. Recognition module -- 5.4.3.1. Immune network model -- 5.4.4. Strategy generation module -- 5.4.5. Decision module -- 5.5. Experimental results -- 5.5.1. Can the agent adapt to a new opponent? -- 5.5.2. Can the agent adapt to consecutive opponents? -- 5.5.3. Can the agent remember previous opponents? -- 5.5.4. Results from the IPD competition -- 5.6. Discussion -- 5.7. Conclusions -- References -- Chapter 6 Exponential Smoothed Tit-for-Tat Michael Filzmoser -- 6.1. Introduction -- 6.2. Exponential Smoothed Tit-for-Tat -- 6.2.1. Exponential Smoothing -- 6.2.2. Strategies for Competitions with and without Noise -- 6.3. Tournament Results -- 6.4. Conclusions -- References -- Chapter 7 Opponent Modelling, Evolution, and The Iterated Prisoner's Dilemma Philip Hingston, Dan Dyer, Luigi Barone, Tim French and Graham Kendall -- 7.1. Introduction -- 7.1.1. Opponent modelling -- 7.1.2. Modeller, the competition entry -- 7.1.3. Anatomy of the modeller -- 7.1.4. Competition performance -- 7.2. Opponent Modelling Versus Evolution -- 7.2.1. The new experiments -- 7.3. Conclusions -- References -- Chapter 8 On Some Winning Strategies for the Iterated Prisoner's Dilemma Wolfgang Slany and Wolfgang Kienreich -- 8.1. Introduction -- 8.2. Analysis of the Tournament Results -- 8.2.1. 2004 competition, league 1 (standard IPD rules, with 223 participating strategies) -- 8.2.2. 2004 competition, league 2 (uncertainty IPD vari- ant, same 223 participating strategies as in the first league).

8.2.3. 2005 competition, league 1 (standard IPD rules, with 192 participating strategies) -- 8.2.4. 2005 competition, league 4 (standard IPD rules, but only non-group, individual strategies were allowed to participate -- 50 participating strategies) -- 8.2.5. Analysis of OmegaTitForTat's (OTFT) performance -- 8.2.6. The practical difficulty of detecting collusion -- 8.3. Details of Our Strategies -- 8.3.1. OmegaTitForTat, or Mr. Nice Guy meets the iterated prisoner's dilemma -- 8.3.1.1. Suspicion -- 8.3.1.2. Randomness -- 8.3.1.3. Exploits -- 8.3.1.4. OTFT -- 8.3.1.5. Examples -- 8.3.1.6. OTFT's behaviour laid bare -- 8.3.2. Our group strategies -- 8.3.2.1. The CosaNostra group strategy, or Organized crime meets the iterated prisoner's dilemma -- 8.3.2.2. The gory details of the CosaNostra group strategy -- 8.3.2.3. TheEmperorAndHisCloneWarriors -- 8.3.2.4. The StealthCollusion group strategy -- 8.4. Analysis of the Performance of the Strategies -- 8.4.1. OmegaTitForTat -- 8.4.2. Group strategies -- 8.4.3. Collusion detection is an undecidable problem -- 8.5. Conclusion -- Acknowledgments -- References -- Chapter 9 Error-Correcting Codes for Team Coordination within a Noisy Iterated Prisoner's Dilemma Tournament Alex Rogers, Rajdeep K. Dash, Sarvapali D. Ramchurn, Perukrishnen Vytelingum and Nicholas R. Jennings -- 9.1. Introduction -- 9.2. The Iterated Prisoner's Dilemma and Related Work -- 9.3. Team Players -- 9.4. Experimental Results -- 9.5. Analysis -- 9.6. Error Correcting Codes -- 9.7. Competition Entry -- 9.8. Conclusions -- A.1. Test Population -- References -- Chapter 10 Is it Accidental or Intentional? A Symbolic Approach to the Noisy Iterated Prisoner's Dilemma Tsz-Chiu Au and Dana Nau -- 10.1. Introduction -- 10.2. Motivation and Approach -- 10.3. Iterated Prisoner's Dilemma with Noise.

10.4. Strategies, Policies, and Hypothesized Policies -- 10.5. Derived Belief Strategy -- 10.6. Learning Hypothesized Policies in Noisy Environ- ments -- 10.6.1. Learning by Discounted Frequencies -- 10.6.2. Deficiencies of Discounted Frequencies in Noisy En- vironments -- 10.6.3. Identifying Deterministic Rules Using Induction -- 10.6.4. Symbolic Noise Detection and Temporary Tolerance -- 10.6.5. Coping with Ignorance of the Other Player's New Behavior -- 10.7. The Move Generator in DBS -- 10.8. Competition Results -- 10.8.1. Overall Average Scores -- 10.8.2. DBS versus the Master-and-Slaves Strategies -- 10.8.3. A comparison between DBSz, TFT, and TFTT -- 10.9. Related Work -- 10.10. Summary and Future Work -- References.