Warning: file_put_contents(aCache/aDaily/post/ai_newz/-2428" target="_blank" rel="noopener" onclick="return confirm('Open this link?\n\n'+this.href);">часть 1</a>, <a href="https://t.me/ai_newz/2428" target="_blank" rel="noopener" onclick="return confirm('Open this link?\n\n'+this.href);">часть 2</a><br/>- <a href="https://t.me/ai_newz/1911" target="_blank" rel="noopener" onclick="return confirm('Open this link?\n\n'+this.href);">Моя любимая статья по диффузионным моделям</a> (база)<br/>- <a href="https://t.me/ai_newz/2332-): Failed to open stream: No such file or directory in /var/www/group-telegram/post.php on line 50
эйай ньюз | Telegram Webview: ai_newz/3478 -
Telegram Group & Telegram Channel
CS492(D): Diffusion Models and Their Applications

Курс по Диффузионным моделям от KAIST (Южная Корея) - если хочется поднять базу, не ходя в университет.

Читает леции вот этот чувак - Minhyuk Sung. На сайте есть записи всех лекций и слайды, плюс 2 гостевые лекции от крутых ученых.

Список лекций:
1 - Course Introduction
2 - Introduction to Generative Models / GAN / VAE
3 - DDPM 1
4 - DDPM 2
5 - DDIM 1
6 - DDIM 2 / CFG
7 - CFG / Latent Diffusion / ControlNet / LoRA
8 - Zero-Shot Applications
9 - Guest Lecture 1 by Or Patashnik
10 - DDIM Inversion / Score Distillation 1
11 - Score Distillation 2
12 - Diffusion Synchronization
13 - Inverse Problems 1
14 - Inverse Problems 2
15 - Probability Flow ODE / DPM-Solver
16 - Flow Matching 1
17 - Flow Matching 2
18 - Course Summary
19 - Guest Lecture 2 by Jiaming Song, Chief Scientist at Luma AI

https://mhsung.github.io/kaist-cs492d-fall-2024/

Ну а еще почитать про диффузию можно у меня :) Вот пара ссылок:
- Как ускорить диффузию часть 1, часть 2
- Моя любимая статья по диффузионным моделям (база)
- Разбор нашей статьи Cashe Me if You Can по ускорению диффузионок
- И ещё пара туториалов, вот первый и второй

#ликбез

@ai_newz



group-telegram.com/ai_newz/3478
Create:
Last Update:

CS492(D): Diffusion Models and Their Applications

Курс по Диффузионным моделям от KAIST (Южная Корея) - если хочется поднять базу, не ходя в университет.

Читает леции вот этот чувак - Minhyuk Sung. На сайте есть записи всех лекций и слайды, плюс 2 гостевые лекции от крутых ученых.

Список лекций:
1 - Course Introduction
2 - Introduction to Generative Models / GAN / VAE
3 - DDPM 1
4 - DDPM 2
5 - DDIM 1
6 - DDIM 2 / CFG
7 - CFG / Latent Diffusion / ControlNet / LoRA
8 - Zero-Shot Applications
9 - Guest Lecture 1 by Or Patashnik
10 - DDIM Inversion / Score Distillation 1
11 - Score Distillation 2
12 - Diffusion Synchronization
13 - Inverse Problems 1
14 - Inverse Problems 2
15 - Probability Flow ODE / DPM-Solver
16 - Flow Matching 1
17 - Flow Matching 2
18 - Course Summary
19 - Guest Lecture 2 by Jiaming Song, Chief Scientist at Luma AI

https://mhsung.github.io/kaist-cs492d-fall-2024/

Ну а еще почитать про диффузию можно у меня :) Вот пара ссылок:
- Как ускорить диффузию часть 1, часть 2
- Моя любимая статья по диффузионным моделям (база)
- Разбор нашей статьи Cashe Me if You Can по ускорению диффузионок
- И ещё пара туториалов, вот первый и второй

#ликбез

@ai_newz

BY эйай ньюз




Share with your friend now:
group-telegram.com/ai_newz/3478

View MORE
Open in Telegram


Telegram | DID YOU KNOW?

Date: |

Emerson Brooking, a disinformation expert at the Atlantic Council's Digital Forensic Research Lab, said: "Back in the Wild West period of content moderation, like 2014 or 2015, maybe they could have gotten away with it, but it stands in marked contrast with how other companies run themselves today." Since January 2022, the SC has received a total of 47 complaints and enquiries on illegal investment schemes promoted through Telegram. These fraudulent schemes offer non-existent investment opportunities, promising very attractive and risk-free returns within a short span of time. They commonly offer unrealistic returns of as high as 1,000% within 24 hours or even within a few hours. "He has to start being more proactive and to find a real solution to this situation, not stay in standby without interfering. It's a very irresponsible position from the owner of Telegram," she said. Pavel Durov, a billionaire who embraces an all-black wardrobe and is often compared to the character Neo from "the Matrix," funds Telegram through his personal wealth and debt financing. And despite being one of the world's most popular tech companies, Telegram reportedly has only about 30 employees who defer to Durov for most major decisions about the platform. Oleksandra Matviichuk, a Kyiv-based lawyer and head of the Center for Civil Liberties, called Durov’s position "very weak," and urged concrete improvements.
from it


Telegram эйай ньюз
FROM American