1. Mastergoal Machine Learning Environment Phase 1 Completion Assessment MSE Project Kansas State University Alejandro Alliana

2. Deliverables Vision Document v 0.1.0 Project Plan 0.1.0 Software Quality Assurance Plan 0.1.0 Prototype

3. Mastergoal Board game with discrete states. Played at different levels. High branching factor. New in AI research.

4. Project Goals Provide an environment to create, repeat, save experiments for creating strategies for playing Mastergoal using ML techniques. Try different AI techniques in the environment of the game

5. Background Traditional approaches –Search in the state space S applying actions A(s t ) to the states and Evaluating the generated states s t+1 using a hand crafted evaluation function Reinforcement Learning –Unsupervised learning. –Temporal difference learning. –Successful with Backgammon. –Problems with some games such as Chess and Go. –TD-Leaf, TD(μ)

6. Risks Inexperience with some algorithms and programming language Exploration vs. exploitation Computational Cost of Evaluation Functions Quality

7. Prototypes demonstration

8. Constraints Export strategies to be used in the Mastergoal plugin environment. CPP programming language

9. Requirements Experiment Management Training strategy Export Strategy Explore game

10. System Components

11. Experiment Management

12. Other Use Cases

13. Documentation standards UML Diagrams Scenario description Coding Standards following the C++ standards Commentary standards following Code Conventions for the Java Programming Language.

14. Testing Standards Unit testing –CppTest Component testing Integration Testing Performance Testing Testing plan

15. Version Control SVN Repository Maven directory Structure standard Tortoise SVN Client

16. Tools IDE –Microsoft Visual Studio Modeling –Rational Rose –Gliffy.com Documentation –Microsoft Word Code control –Tortoise SVN Managing –Process Dashboard –Microsoft Project

17. Cost Estimate COCOMO COCOMO II Use case points

18. COCOMO Effort = 3.2 EAF (Size) 1.05 Time = 2.5 (Effort) 0.38 Where: –Effort is the number of staff months –EAF is the product of 15 effort adjustment factors. –Size is the number of delivered source instructions in KLOC.

19. Cocomo – Effort Adjustment factors IdEffort Adjustment FactorParameter Range Potential ImpactValue SelectedReasoning RELYRequired reliability0.75 - 1.401.871.00Nominal - The application is reliability is not critical DATADatabase size0.94 - 1.161.231.00Nominal -Database access to store games CPLXProduct complexity0.70 – 1.652.361.15High – Product contains reinforcement learning algorithms. TIMEExecution time constraint1.00 – 1.661.661.11High – Experiments must not take too long, since the application is already computationally intensive. STORMain Storage Constraint1.00 – 1.561.561.00Nominal VIRTVirtual machine volatility0.87 – 1.301.491.00Nominal TURNComputer turnaround time0.87 – 1.151.321.00Nominal ACAPAnalyst capability1.46 – 0.712.060.86High. Developer has adequate experience. AEXPApplications experience1.29 – 0.821.571.10Low. Some of the components are new to the developer PCAPProgrammer capability1.42 – 0.702.031.00Nominal VEXPVirtual machine experience1.21 – 0.901.341.00Nominal. Developer has adequate experience with OS systems and tools. LEXPLanguage experience1.14 – 0.951.201.07Low. Developer is new to the C++ language. MODPUse of modern practices1.24 – 0.821.510.91High. The process will follow modern practices. TOOLUse of software tools1.24 – 0.831.490.91High. SCEDRequired development schedule1.23 – 1.101.231.10Low. Project is on a constrained schedule

20. COCOMO Estimate Estimated KLOC (7.5) Effort = 3.2 (1.18) (7.5) 1.05 Effort= 31.32 staff months Time = 2.5 (Effort) 0.38 Time = 9.25 months

21. COCOMO II COCOMO II defines three models for cost estimation: –Applications composition model –Early design model –Post-Architecture model.

22. Application Composition Model Assess Object-Counts Classify each object instance into simple, medium and difficult and weight them. Determine Object-Points Estimate percentage of reuse Determine a productivity rate Compute the estimated person-months

23. Application Composition Model PM = 39/7 = 5.57 Person months –(2.25 ~ 11.07 months)

24. Early Design Model Effort = 2.45 EArch (Size) P Where: –Effort = number of staff-moths –EArch = is the product of seven early design effort adjustment factors –Size = number of function points or KLOC –P are the scaling factors.

25. Post Architecture model Effort = 2.45 (Eapp) (Size) P –Effort = number of staff-moths –EArch = is the product of seventeen post architecture effort adjustment factors –Size = number of function points or KLOC –P = process exponent, same as the early design model. Effort = 33.99 staff months Time = 9.54 months (7.632 ~ 11.93)

26. Project Schedule

27. Phase Two Deliverables Vision document Project Plan Test Plan Architecture Design Formal Requirements Specification Formal Technical Inspection Executable Architecture Prototype.

28. Questions End of presentation

29. Application Composition Model ObjectClassificationObject Points Main screenMedium2 Game exploration screenMedium2 Export strategy screenSimple1 Experiments status reportSimple2 Explore game screenMedium2 Game componentComplex10 Search componentComplex10 Learning componentComplex10 TOTAL39

30. Scaling factors Scale FactorAbbreviationValue. PrecedentednessPRECNominal – 3 Development FlexibilityFLEXLow – 4 Architecture risk resolutionRESLNominal – 3 Team cohesionTEAMExtra high – 0

31. Frameworks Studied Knight Cap Neuro Draugths RL Glue