Arvutiteaduse instituut
  1. Kursused
  2. 2018/19 kevad
  3. Erikursus masinõppes: Stiimulõpe närvivõrkudega (MTAT.03.317)
EN
Logi sisse

Erikursus masinõppes: Stiimulõpe närvivõrkudega 2018/19 kevad

  • Main
  • Timetable
  • Homeworks
  • Projects
  • Links

Timetable

Fall 2018

2018-09-06Lecture 1: Introduction and Course Overview (slides) (test)(watch in the class together)
2018-09-13Lecture 2: Supervised Learning and Imitation (slides) (test)(watch home, test and discussion in class)
2018-09-20Lecture 3: TensorFlow and Neural Nets Review Session
(slides) (notebook)
(watch home if needed, no test)
 Homework 1: Imitation Learning (code)(solution presentations, deadline next day)
2018-09-27Lecture 4: Reinforcement Learning Introduction (slides) (test)(watch home, test and discussion in class)
2018-10-04Lecture 5: Policy Gradients Introduction (slides) (test)(watch home, test and discussion in class)
2018-10-11Homework 2: Policy Gradients (code)(solution presentations 1-5, deadline next day)
2018-10-18Lecture 6: Actor-Critic Introduction (slides) (test)(watch home, test and discussion in class)
2018-10-25Homework 2: Policy Gradients (code)(solution presentations 6-8, deadline next day)
2018-11-01Lecture 7: Value Functions and Q-Learning (slides) (test)(watch home, test and discussion in class)
2018-11-08Lecture 8: Advanced Q-Learning Algorithms (slides) (test)(watch home, test and discussion in class)
2018-11-15Homework 3: Q-Learning and Actor-Critic (code)(solution presentations, deadline next day)
2018-11-22Lecture 9: Advanced Policy Gradients (slides) (test)(watch home, test and discussion in class)
2018-11-29Lecture 10: Optimal Control and Planning (slides) (test)(watch home, test and discussion in class)
2018-12-06Lecture 11: Model-Based Reinforcement Learning (slides) (test)(watch home, test and discussion in class)
2018-12-13Homework 4: Model-Based RL (code)(solution presentations, deadline next day)
2018-12-20Lecture 12: Advanced Model-Based Reinforcement Learning (slides)(watch home, test and discussion in class)
???Lecture 13: Model-Based RL and Policy Learning (slides)(watch home, test and discussion in class)

Spring 2019

2019-02-12Lecture 14: Variational Inference and Generative Models (slides) (test)(watch home, test and discussion in class)
2019-02-19Lecture 15: Reframing Control as an Inference Problem (slides) (test)(watch home, test and discussion in class)
2019-02-26Lecture 16: Inverse Reinforcement Learning (slides) (test)(watch home, test and discussion in class)
2019-03-05Project milestone 1(what task/environment, what is observation space, what is action space, what is reward)
 Homework 5b: Advanced Topics - Soft Actor-Critic (code)(solution presentations, deadline next day)
2019-03-12Lecture 17: Exploration: Part 1 (slides) (test)(watch home, test and discussion in class)
2019-03-19Lecture 18: Exploration: Part 2 (slides)(watch home, test and discussion in class)
2019-03-26Project milestone 2(what algorithm? what codebase? what infrastructure for training?)
 Homework 5a: Advanced Topics - Exploration (code)(solution presentations, deadline next day)
2019-04-02Lecture 19: Transfer and Multi-Task Learning (slides) (test)(watch home, test and discussion in class)
2019-04-09Lecture 20: Meta Reinforcement Learning by Chelsea Finn (slides) (test)(watch home, test and discussion in class)
2019-04-16Lecture 21: Distributed RL by Richard Liaw & Eric Liang (slides) (slides2) (test)(watch home, test and discussion in class)
2019-04-23Project milestone 3(initial results, what improvements over initial results you plan?)
 Homework 5c: Advanced Topics - Meta-Learning (code)(solution presentations, deadline next day)
2019-04-30Lecture 22: Challenges in Deep Reinforcement Learning (test)(watch home, test and discussion in class)
2019-05-07Guest Lecture: Reinforcement learning for Recommender Systems: Some Foundational and Practical Issues by Craig Boutilier(watch home or in class, no test)
2019-05-14Guest Lecture: Real-World Robot Learning:Safety and Flexibility by Gregory Kahn (slides)(watch home or in class, no test)
2019-05-21Guest Lecture: AutoML: Automated Machine Learning by Barret Zoph & Quoc Le (slides)(watch home or in class, no test)
2019-05-28Project presentations
  • Arvutiteaduse instituut
  • Loodus- ja täppisteaduste valdkond
  • Tartu Ülikool
Tehniliste probleemide või küsimuste korral kirjuta:

Kursuse sisu ja korralduslike küsimustega pöörduge kursuse korraldajate poole.
Õppematerjalide varalised autoriõigused kuuluvad Tartu Ülikoolile. Õppematerjalide kasutamine on lubatud autoriõiguse seaduses ettenähtud teose vaba kasutamise eesmärkidel ja tingimustel. Õppematerjalide kasutamisel on kasutaja kohustatud viitama õppematerjalide autorile.
Õppematerjalide kasutamine muudel eesmärkidel on lubatud ainult Tartu Ülikooli eelneval kirjalikul nõusolekul.
Tartu Ülikooli arvutiteaduse instituudi kursuste läbiviimist toetavad järgmised programmid:
euroopa sotsiaalfondi logo