My Hobby: Reviewing and Curating Outstanding EECS Open Courses
One of my hobbies is reviewing high-quality EECS open courses and re-organizing the materials for the most important ones in my own words. I find that distilling the key concepts and structuring them in a learner-friendly way not only deepens my own understanding, but also helps others benefit from these world-class resources.
If you are looking for excellent deep learning CS courses, I recommend checking out this curated course list that I have compiled.
Below are my reorganized materials and notes for several significant courses I have studied.
Autonomous Marine Intelligent Systems
Norwegian University of Science and Technology — TTK4190: Guidance, Navigation and Control of Marine Craft, Aircraft and Drones
This course covers guidance, navigation, and control principles for marine craft. In this course, I focus particularly on the derivation of the kinematic and kinetic equations for underwater unmanned systems
\[M\dot{\nu}+C(\nu )\nu +D(\nu )\nu +g(\eta )=\tau\]which form the foundation for precise attitude control, path planning, and coordinated formation.
Deep Learning
MIT 6.S898: Deep Learning
A cutting-edge deep learning course covering architectures, optimization techniques, and applications. My notes focus on intuitive explanations of core concepts like attention mechanisms, representation learning, and large-scale model training.
Reinforcement Learning
UCB CS 285: Deep Reinforcement Learning
An in-depth exploration of policy optimization, value-based methods, and model-based RL. I mainly reorganized the sections on Policy Gradient, Actor-Critic Algorithms, and Value Function Methods.
- lecture 4: Intro to RL
- lecture 5: Policy Gradient
- lecture 6: Actor-Critic Algorithm
- lecture 7: Value Function Methods
Edge Computing
MIT 6.S965: TinyML and Efficient Deep Learning Computing
This course focuses on deploying deep learning models in resource-constrained environments. My summary includes hands-on notes on the principles of common convolution operators and neural network efficiency metrics, and will be extended in the future to cover quantization, pruning, NAS, and distillation.
Distributed Systems
MIT 6.824: Distributed Systems
A legendary course on distributed computing principles, from consensus protocols (like Raft) to fault tolerance. I emphasize diagrams and execution traces to make complex concepts approachable.
- Lecture 1: Mapreduce
- Lecture 2: Go,RPC,Threads
- A tour of Go
- Lab 1: Re-implement MapReduce model described in OSDI’ 04
Computer Science Fundamentals
-
UCB CS61A: Structure and Interpretation of Computer Programs
Focused on computational thinking and functional programming fundamentals.
-
UCB CS61B: Data Structures and Algorithms
A comprehensive treatment of core data structures, algorithm analysis, and implementation patterns.
Research Contributions
Beyond course reviews, I have also created several awesome reading lists focused on multivariate time series research. These lists have collectively received 500+ GitHub stars, reflecting the value they bring to the research community.
- Awesome-time-series-segmentation-papers, (post link), [500+ GitHub stars!]
- Awesome-multivariate-time-series-anomaly-detection-algorithms
- Awesome-multivariate-time-series-clustering-algorithms
If you are passionate about self-learning in EECS, I hope my reorganized course materials and curated resources can serve as useful study companions.