Hung-yi Lee

439K subscribers

李宏毅 webpage for ML 2021: https://speech.ee.ntu.edu.tw/~hylee/ml/2021-spring.html

1:03:45

AI 要跨越盧比孔河了嗎?自我成長的 AI 離我們多遠 (上集)

Hung-yi Lee · 47K views · 4 months ago

1:32:21

Harness Engineering: Sometimes Language Models Aren't Unintelligent, They Just Lack Proper Human ...

Hung-yi Lee · 158K views · 5 months ago

56:42

[ML 2021 (English version)] Lecture 9: Convolutional Neural Networks

Hung-yi Lee · 7.8K views · 5 years ago

11:01

[INTERSPEECH 2020] Semi-supervised Learning for Multi-speaker Text-to-speech Synthesis

Hung-yi Lee · 1.6K views · 5 years ago

2:00:11

【生成式人工智慧與機器學習導論2025】第 8 講:通用模型的終身學習 (Fine-tuning, Model Editing, Model Merging, Test-Time Training)

Hung-yi Lee · 34K views · 10 months ago

1:30:51

ML Lecture 21-2: Recurrent Neural Network (Part II)

Hung-yi Lee · 139K views · 9 years ago

22:29

AI Agent (2/3): AI Agent 之間可以有什麼樣的互動

Hung-yi Lee · 50K views · 6 months ago

1:09:09

Is AI Crossing the Rubicon? How Far Are We from Self-Improving AI? (Part 2)

Hung-yi Lee · 27K views · 4 months ago

53:14

[ML 2021 (English version)] Lecture 1: Predicting the views of this channel - ML Introduction (1/2)

Hung-yi Lee · 71K views · 5 years ago

41:34

DRL Lecture 2: Proximal Policy Optimization (PPO)

Hung-yi Lee · 107K views · 8 years ago

1:18:35

ML Lecture 1: Regression - Case Study

Hung-yi Lee · 480K views · 9 years ago

21:11

[DLHLP 2020] Deep Learning for Dependency Parsing

Hung-yi Lee · 2.9K views · 6 years ago

28:18

【機器學習2021】自注意力機制 (Self-attention) (上)

Hung-yi Lee · 335K views · 5 years ago

24:44

SUPERB: Is self-supervised learning universal in speech processing tasks? (English version)

Hung-yi Lee · 5.5K views · 4 years ago

2:04:44

【生成式人工智慧與機器學習導論2025】第3講:解剖大型語言模型

Hung-yi Lee · 124K views · 1 year ago

1:29:43

如何讓 Transformer 知道輸入 Token 的順序?Absolute、Relative、RoPE、到沒有 Positional Embedding

Hung-yi Lee · 26K views · 6 months ago

YouTube, with the door locked.

Aegis plays a clean stream instead of YouTube's player, so pre-roll ads, trackers, and fingerprinting never ride along. Drop Shields any time if you want the official player back.