Flowers

DataHour: Diffusion Models for Generative Arts

Online 21-10-2022 07:00 PM to 21-10-2022 08:00 PM
  • 3542

    Registered

  • Knowledge and Learning

    Prizes

DataHour Recording

Find the resources used in the DataHour HERE.

About the DataHour:

Diffusion models form a separate class of Adversarial Networks aimed at diffusing different latent data points through a space in order to create a different latent space. Generally traditional GAN variants like DCGAN/CycleGAN etc are predominantly used for generating different latent spaces, diffusion however is a dual step process which tries to merge different latent spaces to create an entirely new one.

In this DataHour, Abhilash will explain about the flow of generative latent state representation from GANs, VAEs to Diffusion methods. Since Diffusions are based on the Markov model, he will be building small diffusion models to make you understand latent space representation from images. He will also demonstrate to you how to analyze contemporary multimodal models such as Dall-e/CLIP/unCLIP/GLIDE/Imagen in the context of Diffusion models to replicate and create "Generative AI" which is taking the NFT world by storm.


Prerequisites:

Basic Familiarity with GANs concept (or 0-sum game theory), convergence/divergence principles, linear algebra, torch/tensorflow intermediate level and passion of learning Data Science.


Who is this DataHour for?

  • Students & Freshers who want to build a career in the Data-tech domain.
  • Working professionals who want to transition to the Data-tech domain.
  • Data science professionals who want to accelerate their career growth


Speaker:

Abhilash Majumder, Senior SuperComputing Engineer at Intel

Abhilash Majumder is currently working as a senior SuperComputing Engineer at Intel enabling next generation GPUs for exa scale computing, deep learning of very large models spanning vision and multimodal contexts. He was a research scientist for Morgan Stanley and a collaborator with Imperial College London for finetuning and building large language models . Prior to that, he was a part of HSBC working on knowledge graphs, semantic bots and transformers, part of Unity Technologies (ML Agents and Reinforcement Learning) ; and a part of Google Research for the Albert model. He is also an author, mentor and have provided Deep Learning/Quantum Deep Learning related sessions at Python conferences globally. 


Connect with Abhilash on linkedin , twitter or Github

ALT

Please register/login to participate in the contest

Please register to participate in the contest

Please register to participate in the contest

Closed

Support

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy.

Feedback

We believe in making Analytics Vidhya the best experience possible for Data Science enthusiasts. Help us by providing valuable Feedback.