C programming language implementation of aima-pseudocode from Russell And Norvig's "Artificial Intelligence - A Modern Approach, Fourth Edition".
Why C? It is the best "get it done" computer programming language.
C is a small programming language. The first edition book is 236 pages. The second edition book is 288 pages. The "reference manual" section is only 35 pages and 49 pages respectively.
Then why is C considered hard? The usual answer is pointers. The real answer is choice. C only provides portable access to general processor operations on memory. The language has no support for dynamic memory or I/O (input / output). C callable routines (e.g. the C standard library) must be provided to enable C programs to interface with an environment. Everything else is programmer choice.
For example, there are a few conventions used in the book that neither C nor the standard library address:
- From the preface, "The main unifying theme is the idea of an intelligent agent."
- From Appendix B "notes on languages and algorithms", the concepts "persistent variables" and "yield".
These concurrent styles can be implemented using various coroutine techniques but don't enable parallelism. For both concurrency and parallelism POSIX Threads is used. To reduce the complexity of using threads a channel style is used using pthreadChannel. These choices provide high level clarity and low level performance.
If there is nothing in the "C implementation" column, I haven't gotten that far.
Index of implemented algorithms
| Pseudo-code Algorithm | Prototype | Implementation |
|---|---|---|
| TABLE-DRIVEN-AGENT | tableDrivenAgent.h | tableDrivenAgent.c |
| REFLEX-VACUUM-AGENT | reflexVacuumAgent.h | reflexVacuumAgent.c |
| SIMPLE-REFLEX-AGENT | simpleReflexAgent.h | simpleReflexAgent.c |
| MODEL-BASED-REFLEX-AGENT | modelBasedReflexAgent.h | modelBasedReflexAgent.c |
| BEST-FIRST-SEARCH | ||
| BREADTH-FIRST-SEARCH | ||
| UNIFORM-COST-SEARCH | ||
| ITERATIVE-DEEPENING-SEARCH | ||
| DEPTH-LIMITED-SEARCH | ||
| BIBF-SEARCH | ||
| RECURSIVE-BEST-FIRST-SEARCH | ||
| HILL-CLIMBING | ||
| SIMULATED-ANNEALING | ||
| GENETIC-ALGORITHM | ||
| AND-OR-SEARCH | ||
| ONLINE-DFS-AGENT | ||
| LRTA*-AGENT | ||
| MINIMAX-SEARCH | ||
| ALPHA-BETA-SEARCH | ||
| MONTE-CARLO-TREE-SEARCH | ||
| AC-3 | ||
| BACKTRACKING-SEARCH | ||
| MIN-CONFLICTS | ||
| TREE-CSP-SOLVER | ||
| KB-AGENT | ||
| TT-ENTAILS | ||
| PL-RESOLUTION | ||
| PL-FC-ENTAILS? | ||
| DPLL-SATISFIABLE? | ||
| WALKSAT | ||
| HYBRID-WUMPUS-AGENT | ||
| SATPLAN | ||
| UNIFY | ||
| FOL-FC-ASK | ||
| FOL-BC-ASK | ||
| APPEND | ||
| AIR-CARGO-TRANSPORT-PROBLEM | ||
| SPARE-TIRE-PROBLEM | ||
| BLOCKS-WORLD | ||
| HAVE-CAKE-AND-EAT-CAKE-TOO-PROBLEM | ||
| GRAPHPLAN | ||
| REFINEMENT-HIGH-LEVEL-ACTIONS | ||
| HIERARCHICAL-SEARCH | ||
| ANGELIC-SEARCH | ||
| JOB-SHOP-SCHEDULING-PROBLEM | ||
| DT-AGENT | ||
| ENUMERATION-ASK | ||
| ELIMINATION-ASK | ||
| PRIOR-SAMPLE | ||
| REJECTION-SAMPLING | ||
| LIKELIHOOD-WEIGHTING | ||
| GIBBS-ASK | ||
| FORWARD-BACKWARD | ||
| FIXED-LAG-SMOOTHING | ||
| PARTICLE-FILTERING | ||
| OUPM | ||
| NET-VISA | ||
| RADAR | ||
| GENERATE-IMAGE | ||
| GENERATE-MARKOV-LETTERS | ||
| INFORMATION-GATHERING-AGENT | ||
| VALUE-ITERATION | ||
| POLICY-ITERATION | ||
| POMDP-VALUE-ITERATION | ||
| DOUBLES-TENNIS-PROBLEM | ||
| LEARN-DECISION-TREE | ||
| CROSS-VALIDATION-WRAPPER | ||
| DECISION-LIST-LEARNING | ||
| ADABOOST | ||
| CURRENT-BEST-LEARNING | ||
| VERSION-SPACE-LEARNING | ||
| MINIMAL-CONSISTENT-DET | ||
| FOIL | ||
| PASSIVE-ADP-AGENT | ||
| PASSIVE-TD-AGENT | ||
| Q-LEARNING-AGENT | ||
| HITS | ||
| CYK-PARSE | ||
| SENTENCE-TREE | ||
| MONTE-CARLO-LOCALIZATION | ||
| POWERS-OF-2 |