Lu, Yukuan

Resources for EE(Electrical Engineering) learning, including signal processing, information theory, etc.

Resources for essential topics of Machine Learning and Deep learning, including Natural language processing (NLP), Computer Vision (CV), Reinforcement Learning (RL), Self-Supervised Learning (SSL), etc.

Sources:

1. James McClellan, Ronald Schafer & Mark Yoder. (2015). IIR Filters. DSP First (2nd ed., pp. 394-457). Pearson.

Common problems of Shannon entropy in information theory.

Sources:

1. The spelled-out intro to neural networks and backpropagation: building micrograd
2. Stanford CS231N, Lecture 4

Other useful resources:

1. StatQuest's Neural Networks videos

2. Efficient backprop. You need do download it via

   1	wget http://yann.lecun.com/exdb/publis/pdf/lecun-98b.pdf

The python script and jupyter notebook used in this article can be found at here.

This article is a step-by-step explanation of neural networks which are extensively used in machine learning. It only involves the most basic case where the input of a neuron is a vector (1-D tensor)and output of a neuron is a scalar (0-D tensor). But the idea holds for higher, please reter to Derivatives, Backpropagation, and Vectorization for details.

Ref:

1. EE 376A: Information Theory. Winter 2018. Lecture 6. - Stanford University
2. EE 376A: Information Theory. Winter 2017. Lecture 4. - Stanford University
3. Elements of Information Theory

Sources:

1. James McClellan, Ronald Schafer & Mark Yoder. (2015). FIR Filters. DSP First (2nd ed., pp. 147-193). Pearson.

Sources:

1. Jiames McClellan, Ronald Schafer & Mark Yoder. (2015). Sinusoids. DSP First (2nd ed., pp. 9-40). Pearson.

Sources:

1. James McClellan, Ronald Schafer & Mark Yoder. (2015). Sinusoids. DSP First (2nd ed., pp. 9-40). Pearson.

Sources:

1. World Models 2018 paper
   • Training doc
2. Mastering Diverse Domains through World Models, aka Dreamer V3.
3. A Generalist Agent.
4. Learning General World Models in a Handful of Reward-Free Deployments.
5. Learning General World Models in a Handful of Reward-Free Deployments.