Evolutionary Computation
אלגוריתמים אבולוציוניים
Lecturer: Prof. Moshe Sipper
202-2-5651, 4 credits
Administrative Details
- Prerequisites: Automata, Systems Programming, Algorithms, PPL
- Highly recommended prerequisite: Python programming
- Credits: 4
- Grade:
- 18%: Midterm 1
- 19%: Midterm 2
- 29%: Presentation
- 34%: Project
- You must pass all above 4 course components in order to pass the course.
- Midterm:
- If you miss a midterm due to a valid reason according to the university regulations (see Section 7.2), then you will take an oral makeup exam at the lecturer’s office, at a date and time decided by the lecturer.
- If you miss a midterm due to an invalid reason then the midterm’s grade will count as 0.
- Sample midterm questions.
- Presentation:
- Each student will present, on their own, a topic/paper from the research literature.
- The presentation topic/paper must be approved by the lecturer.
- You must make a selection by April 4.
- If you do not make a selection by April 11, 10 points will be taken off the final grade.
- You can pick a time slot through moodle.
- Presentation length: 8-10 minutes.
- Scoring rubric: Organization (6), Knowledge (6), Text (6), Graphics (6), Elocution (4), Eye Contact (1).
- Project:
- The project must be done in pairs or threesomes.
- The topic must be approved by the lecturer.
- A report must be submitted by the end of the semester (June 18).
- The report must include the following seven sections:
- A short introduction of the domain being investigated.
- A description of the problem or phenomenon studied.
- An explanation of the methods and algorithms employed.
- An overview of the software (not a listing of the code).
- An account of the results obtained.
- Some interesting conclusions.
- Bibliographic references.
- Language: English or Hebrew.
- Length: 6-8 pages.
- Don’t include the code in the report.
- Upload the report to the course moodle as a PDF file.
- Links to 3 sample reports: Librarian, Poker, Bomberman (note: these are longer, yours should be 6-8 pages—keep focused).
Class Material
- Evolutionary Computation
- tiny_ga
- NSGA II
- Schema theorem
- GP: Koza, Koza Tutorial, Koza & Poli, Herrmann
- tiny_gp
- Linear GP
- Cartesian GP
- Grammatical Evolution
- Koza’s vids
- SAFE
- OMNIREP
- Novelty search
- Machine Learning
- Decision trees
- Random forests
- Linear regression
- Logistic regression
- (Deep) neural networks
- AdaBoost
- Gradient boosting
- Clustering, k-means
- Dimensionality reduction, PCA