李宏毅GAN学习笔记（1）

GAN Lecture 1

---

Yann LeCun: Adversarial training is the coolest thing since sliced bread.

Outline

• Basic Idea of GAN
• GAN as Structured Learning
• Can Generator Learn by Itself?
• Can Discriminator Generate?
• A Little Bit Theory

Basic Idea of GAN

Generation

Generation: it is a neural network or a function.

input vector: each dimension of input vector represents some characterstics.

Discriminator

Discriminator: it is a neural network or a function.

output scalar: large value means real, smaller value means fake.

GAN Algorithm

---

• Initialize generator and discriminator

• In each training iteration：

  Step 1: fix generator G, and update discriminator D
  

  

  Step 2: fix discriminator D, and update generator G
  

  

---

Initialize for D and for G

• In each training iteration:

  • Sample m examples from database
  • Sample m noise samples from a distrubution
  • Obtaining generated data
  • Update discriminator parameters to maximize

  Learning D

  • Sample m noise samples from a distribution
  • Update generator parameters to maxmize

  Learning G

GAN as Structured Learning

Structured Learning Approach

• Bottom Up: Learn to generate the object at the component level

  缺乏大局观

• Top Down: Evaluating the whole object, and find the best one

Can Generator Learn by Itself?

若使用Auto-Encoder技术作为Generator，那么需要更深的网络结构。

Can Discriminator Generate?

Discriminator -- Training

• General Algorithm
  • Given a set of positive example, randomly generate a set of negative examples.
• In each iteration
  • Learn a discriminator D that can discriminate positive and negative examples.
  • Generate negative examples by discriminator D:

Generator v.s. Discriminator

• Generator
  • Pros:
    • Easy to generate even with deep model.
  • Cons:
    • Imitate the appearance.
    • Hard to learn the correlation between components.

---

• Discriminator
  • Pros:
    • Considering the big picture.
  • Cons:
    • Generation is not always feasible, especially when your model is deep.
    • How to do negative sampling?

Discriminator + Generator -- Training

• General Algorithm
  • Given a set of positive example, randomly generate a set of negative examples.
• In each iteration
  • Learn a discriminator D that can discriminate positive and negative examples.
  • Generate negative examples by discriminator D: