Lecture 19: Generative Models I

Показать описание

Lecture 19 is the first of two lectures about generative models. We compare supervised and unsupervised learning, and also compare discriminative vs generative models. We discuss autoregressive generative models that explicitly model densities, including PixelRNN and PixelCNN. We discuss autoencoders as a method for unsupervised feature learning, and generalize them to variational autoencoders which are a type of generative model that use variational inference to maximize a lower-bound on the data likelihood.

_________________________________________________________________________________________________

Computer Vision has become ubiquitous in our society, with applications in search, image understanding, apps, mapping, medicine, drones, and self-driving cars. Core to many of these applications are visual recognition tasks such as image classification and object detection. Recent developments in neural network approaches have greatly advanced the performance of these state-of-the-art visual recognition systems. This course is a deep dive into details of neural-network based deep learning methods for computer vision. During this course, students will learn to implement, train and debug their own neural networks and gain a detailed understanding of cutting-edge research in computer vision. We will cover learning algorithms, neural network architectures, and practical engineering tricks for training and fine-tuning networks for visual recognition tasks.

Рекомендации по теме

Комментарии

26:00, "we could ask a model: give me an image of a cat with a purple tail, but i dont think itll work". amazing how within just a few years we have already reached this point where we can synthesize images from arbitrary input.

matato

Justin is the best lecture teacher I've ever seen on Youtube! He can always present the most complicated things in a clear and simple way. Thanks!

antonywill

I am extremely glad that Generative Models were spanned over 2 lectures; excellent lecture as always!

syedhasany

This professor is an excellent communicator.

yearoldman

Extremly thanksful for this lecture, finally getting the intuitions behind generative model. Very valuable thanks again, awesome lecture

ZinzinsIA

i gained a better understanding of generative models as soon as i saw the thumbnail without even watching the video thanks

frommarkham

My left ear enjoyed this lecture a lot :P

AkshayRoyal

For pixelRNN, why not mention the sampling methods (greedy, stochastic, temperature control, and maybe even beam search) which are quite related to the current GPT generation methods. Right?

heejuneAhn

I am quite confused at 1:08, where q(z|x) is posterior of decoder. But actually, we are using encoder to estimate q(z|x). So what is the implication of the terminology here? I'd really appreciate it if anyone can shed some light here.

DED_Search

Can somebody help me with the concept of probability? At 31:44 he talks about how to train a model by a given dataset. It says the goal is to find the value W for a unsupervised model is to maximize the probablility of training data. I am confused with this "probability of training data". Does it mean the probability of when a training data x(i) is given the output to be the same x(i)? Like the cost function of a auto encoder( square of x_hat - x).
My background knowledge is not good enough to look up for papers or math textbooks. so please help me!

kyungryunlee

Can you explain why the Pixel RNN model explicit pdf model? Can you express the function of the pdf of the model? What do you mean by "explicit"? To be explicit the probability should be a form of prob(x1, x2, x3, xn), where xi is the value of each pixel. Can you express it like that? And can you explain how we train the PixelRNN? e.g., the output has a probability of 0 to 255 values, and is L1 or L2 loss applied with the training images?

heejuneAhn

If you just assume z is Gaussian, it really become Gaussian? In principle and general, the latent vector has any distribution. So we have to add one more constraint (the latent should be digonal covariance multivariate Gaussian) to the Autoencoder when we training

heejuneAhn

Lecture 19: Generative Models I

Lecture 19: Generative Models I

Lecture 19-1. Generative Models I

Lecture 19-2. Generative Models I

Lecture 19 - Efficient Video Understanding and Generative Models | MIT 6.S965

Lecture 19 - Efficient Video Understanding and Generative Models | MIT 6.S965

Lecture 19: Generative Models Part 1 (UMich EECS 498-007)

Computational Creativity Lecture 19: Generative Models for Music

Lecture 19-1. Generative Models II

MIT 6.S191 (2024): Deep Generative Modeling

Lecture 20-1. Generative Models II

Lecture 7.2: Generative Models (Multimodal Machine Learning, Carnegie Mellon University)

CS 182: Lecture 19: Part 1: GANs

Stanford CS236: Deep Generative Models I 2023 I Lecture 1 - Introduction

Lecture 18-2. Generative Models I

[Lecture 19] 11785 Intro to Deep Learning - Fall 2018

Machine Learning: Lecture 24b: Discriminative and Generative models

Interpreting Deep Generative Models for Interactive AI Content Creation by Bolei Zhou (CUHK)

AI Course 2021 - week 9 - Generative models

MIT 6.S191 (2019): Deep Generative Modeling

Lecture 19-2. Generative Models II

Generative Models for Image Synthesis

Lecture 19 RL as Inference 1

Deep Generative Models by Thomas Lucas

Giovanni Conforti, Alain Durmus - An introduction to Score-based Generative Models - Lecture 1