TEDx Talks - Chess & Complexity | Adit Sood | TEDxSharjah English School
The video explains that games like tic-tac-toe and checkers are solved games, meaning their outcomes can be predicted if both players play perfectly. Chess, however, is far more complex due to its vast number of possible games, known as Shannon's number, which exceeds the number of atoms in the observable universe. This complexity makes chess an ideal testing ground for AI and algorithms. Modern chess engines like Stockfish use search algorithms and AI to evaluate millions of positions per second, achieving superhuman performance but still relying on approximations rather than solving the game outright. Solving chess would require storing information for every possible position, a task beyond current computational capabilities. Quantum computing, with its ability to process information in parallel and faster than classical computers, holds promise for solving such complex problems. Google's quantum computer chip Willow demonstrated this potential by completing a calculation in minutes that would take classical supercomputers septillions of years. However, quantum computing is still in its early stages and requires further development to reach its full potential. The future of solving chess depends on technological advancements and innovative problem-solving approaches.
Key Points:
- Chess is vastly more complex than tic-tac-toe and checkers, with 10^120 possible games, making it unsolved.
- Modern chess engines like Stockfish achieve superhuman performance but rely on approximations.
- Quantum computing could potentially solve chess by processing information faster and in parallel.
- Google's quantum chip Willow showed promise by performing calculations far beyond classical supercomputers.
- Solving chess depends on future technological advancements and innovative problem-solving.
Details:
1. 🎉 A Curious Start
2. ♟️ Understanding Perfect Information
- Perfect information games are those where all players have complete knowledge of the game state and the actions taken by other players.
- In games like chess, players can theoretically predict the outcome if both play perfectly, highlighting the strategic depth and complexity.
- Tic-tac-toe is a solved game, meaning perfect play results in a draw, illustrating the concept of predictable outcomes with perfect information.
- Checkers was solved in 2007 by Jonathan Schaefer's team using the Chinook program, proving that with perfect play, the game ends in a draw, showcasing the limits of strategic variation in perfect information scenarios.
3. 🔍 The Complexity of Chess
- Chess has 10^120 possible games, known as Shannon's number, which is more than the atoms in the observable universe.
- Each chess position on average has 35 potential moves, leading to exponential growth in possible positions.
- After just four turns, there are almost 85 billion different possible positions in a chess game.
- The complexity of chess makes it an ideal subject for study in mathematics and computer science.
4. 🤖 The Power of Chess Engines
- Modern chess engines use advanced techniques to excel at chess, including generating all legal moves and applying search algorithms and AI to evaluate the best move.
- These engines can evaluate millions of positions per second, utilizing optimization techniques to achieve superhuman performance.
- Stockfish, a leading chess engine, has an estimated rating of 3,642, significantly higher than Magnus Carlsen, one of the world's top-rated chess players.
5. 🔍 Tackling Chess's Complexity
- Modern chess engines rely on approximation rather than solving chess outright, highlighting the complexity of the game.
- Chess endgames have been solved for positions with up to seven pieces using backward induction, generating an 18 terabyte database, demonstrating the scale of computational resources involved.
- Expanding the solved endgame database to include all 32 pieces would require storage many orders of magnitude larger than current capabilities, indicating significant technological limitations.
- The current limitations in storage and computational power suggest that solving chess completely is not feasible with existing technology, impacting the development strategies of modern chess engines.
- These findings underscore the importance of approximation techniques and heuristic methods in advancing the capabilities of chess engines despite unsolved complexities.
6. ⚛️ Quantum Computing's Potential
- Today's fastest supercomputers, capable of quintillions of operations per second, would take longer than the age of the universe to compute certain problems.
- Quantum computing can revolutionize computational capacity by utilizing quantum mechanics, allowing it to solve complex problems beyond the reach of classical computers.
- Quantum bits (qubits) can exist in superposition, representing both 0 and 1 simultaneously, unlike classical bits, enhancing computational speed and efficiency.
- While classical computers process tasks sequentially, quantum computers can perform multiple tasks in parallel, significantly increasing speed.
- Quantum algorithms offer a substantial reduction in computation time for complex evaluations, providing an edge over classical methods.
- Google's quantum computer chip, Willow, performed a calculation in 5 minutes that the fastest supercomputers would take over 10 septillion years to complete, showcasing its immense capability.
- Despite its potential, quantum computing is in early stages of development, with significant research needed to overcome current challenges and fully realize its capabilities.
7. 📈 Insights from Checkers
- Jonathan Schaefer's 16-year project emphasized the importance of recognizing technological advancements.
- The completion of the checkers project in 2007 highlighted the need for long-term commitment to achieve breakthroughs.
- Schaefer's experience underscores the value of patience and persistence in technological research and development.