Deep Learning Course (NYU, Spring 2020) Yann Lecun

Deep Learning course at NYU, Spring 2020. Taught by Yann LeCun & Alfredo Canziani. With practical applications using PyTorch.

图灵奖获得者给你讲深度学习

• 课时1：Week 1 – Lecture- History, motivation, and evolution of Deep Learning (1小时38分56秒)

• 课时2：Week 1 – Practicum- Classification, linear algebra, and visualisation (52分30秒)

• 课时3：Week 2 – Lecture- Stochastic gradient descent and backpropagation (1小时43分16秒)

• 课时4：Week 2 – Practicum- Training a neural network (56分59秒)

• 课时5：Week 3 – Lecture- Convolutional neural networks (1小时38分15秒)

• 课时6：Week 3 – Practicum- Natural signals properties and CNNs (48分21秒)

• 课时7：Week 4 – Practicum- Listening to convolutions (51分1秒)

• 课时8：Week 5 – Lecture- Optimisation (1小时29分5秒)

• 课时9：Week 5 – Practicum- 1D multi-channel convolution and autograd (44分58秒)

• 课时10：Week 6 – Lecture- CNN applications, RNN, and attention (1小时28分47秒)

• 课时11：Week 6 – Practicum- RNN and LSTM architectures (53分33秒)

• 课时12：Week 7 – Practicum- Under- and over-complete autoencoders (55分3秒)

• 课时13：Week 7 – Lecture- Energy based models and self-supervised learning (1小时37分18秒)

• 课时14：Week 8 – Lecture- Contrastive methods and regularised latent variable models (1小时39分25秒)

• 课时15：Week 8 – Practicum- Variational autoencoders (58分4秒)

• 课时16：Week 9 – Lecture- Group sparsity, world model, and generative adversarial networks (GANs) (1小时58分24秒)

• 课时17：Week 9 – Practicum- (Energy-based) Generative adversarial networks (1小时15分11秒)

• 课时18：Week 10 – Practicum- The Truck Backer-Upper (1小时25秒)

• 课时19：Week 10 – Lecture- Self-supervised learning (SSL) in computer vision (CV) (2小时41秒)

• 课时20：Week 11 – Practicum- Prediction and Policy learning Under Uncertainty (PPUU) (1小时23分18秒)

• 课时21：Week 11 – Lecture- PyTorch activation and loss functions (1小时53分44秒)

• 课时22：Week 12 – Practicum- Attention and the Transformer (1小时18分1秒)

• 课时23：Week 12 – Lecture- Deep Learning for Natural Language Processing (NLP) (1小时40分56秒)

• 课时24：Week 13 – Practicum- Graph Convolutional Neural Networks (GCN) (1小时10分1秒)

• 课时25：Week 13 – Lecture- Graph Convolutional Networks (GCNs) (2小时22秒)

• 课时26：Week 14 – Lecture- Structured prediction with energy based models (2小时7分30秒)

• 课时27：Week 14 – Practicum- Overfitting and regularization, and Bayesian neural nets (1小时11分27秒)

• 课时28：Week 15 – Practicum part A- Inference for latent variable energy based models (EBMs) (59分4秒)

• 课时29：Week 15 – Practicum part B- Training latent variable energy based models (EBMs) (58分56秒)

• 课时30：Matrix multiplication, signals, and convolutions (47分1秒)

• 课时31：Supervised and self-supervised transfer learning (with PyTorch Lightning) (1小时11分23秒)

• 课时32：Four decades in Machine Learning- a Personal Journey by Yann LeCun (1小时31分22秒)