defter*
defter / katalog / CS 549
CS 549

Learning for Robotics

Fundamental ideas and techniques for constructing intelligent (robotic) systems acting in the world. Perception (including approaches based on deep learning and approaches based on 3D geometry), planning (robot kinematics and trajectory generation, collision-free motion planning, task-and-motion planning, and planning under uncertainty), as well as dynamics and control (both model-based and learning-based). Deep reinforcement learning approaches (policy gradients, actor-critic algorithms, value function and q-function methods, inverse reinforcement learning, meta-learning). Highlights of state-of-the-art methods and application domains.

Credit3
ECTS5
BölümComputer Engineering
FacultyFaculty of Engineering

Hocalar 1 bu dönem · 0 geçmiş

Bu dönem (2025-2026 Spring) · 1 section
Salih Özgür Öğüz

→ STARS müfredatı / syllabus

Materyal — 0 dosya

Bu derste henüz materyal yok.

İlk dosyayı sen ekleyebilirsin — notlar, geçmiş finaller, çözümler, cheat-sheet, ne varsa. Drive linki / PDF / ZIP / fotoğraf, hepsi olur.

Şu an: mail at, ben düzenleyip yayına alayım. Form/upload UX yakında geliyor (Kimya tasarlıyor).

↑ konuya CS 549 yaz

Müfredat detayı STARS syllabus

📚 Önerilen kaynaklar

  • Önerilen Robotic Manipulation: Perception Planning, and Control · Russ Tedrake · 2023
  • Önerilen Deep Learning Ian Goodfellow, Yoshua Bengio · Aaron Courville · 2016
  • Önerilen Reinforcement Learning: An Introduction Richard S. Sutton, Andrew G. Barto · 2020
  • Önerilen Reinforcement Learning and Optimal Control Dimitri P. Bertsekas · 2023

⚖️ Değerlendirme

  • 30% — Midterm: Mt (×1)
  • 25% — Homework: Hw (×4)
  • 40% — Term project: Pr (×1)
  • 5% — Quiz: Quiz (×5)

⚠️ FZ engelleyen şartlar

There is no final exam for this course, however, any one of the following will directly result in an F grade: (1) not submitting a project or homework (including report), (2) being absent in the midterm, (3) being absent in a project presentation.

🤖 GenAI politikası

We follow the Generative AI policy guideline of Bilkent University which can be found here: https://w3.bilkent.edu.tr/bilkent/generative-artificial-intelligence-genai-guideline/

📅 Haftalık müfredat

Intelligent interaction: Passive vs. Active/Embodied AI Robotics Fundamentals: Kinematics (FK, IK, Diff IK, Optimization) Kinematics + Motion Planning Motion Planning (Trajectory Optimization, Sampling-based Methods) Perception: geometric vision, point clouds + integration with robotic tasks Deep perception: representation learning for robotic manipulation Vision-Language-Action (VLAs) Models Vision-Language-Action (VLAs) Models Vision-Language-Action (VLAs) Models Deep Reinforcement Learning (DRL) Deep Reinforcement Learning (DRL) Sequential Robotic Manipulation, Task and Motion Planning (TAMP) Learning for Task and Motion Planning (TAMP) Final project presentations ECTS - Workload Table: Activities Number Hours Workload Homework 6 8 48 Course hours 14 3 42 Midterm exam 1 2 2 Preparation for Midterm exam 1 16 16 Project (including preparation and presentation if applicable) 1 45 45 Total Workload: 153 Total Workload / 30: 153 / 30 5.1 ECTS Credits of the Course: 5 Type of Course: Lecture - Independent Study - Project Course Material: Lecture Notes - Slides Teaching Methods: Assignment - Presentations