Giter VIP home page Giter VIP logo

deep-learning-theory-paper-discussion's Introduction

Deep Learning Theory Discussion :)

image

The repository contains lists of papers on theoretical deep learning and how relevant techniques are being used to further enhance deep learning. The repository is organized by Amartya Roy, and Anirbit Mukherjee

Important Papers

  • Comparative Generalization Bounds for Deep Neural Networks

    • Amartya's Notes Recent research suggests that deep neural networks are able to generalize well to new data. This paper looks at how corrupted labels affect the extent of intermediate layer NCC separability. Authors have proposed a novel generalization bound that estimates the likelihood that the effective depth of a trained neural network is strictly smaller than the minimal depth required to achieve NCC separability with partially corrupted labels. They have introduced the concept of “effective depth’ in neural networks, which refers to the lowest layer at which the features are nearest class-center separable.
  • Why Can GPT Learn In-Context? Language Models Implicitly Perform Gradient Descent as Meta-Optimizers

    • Amartya's NotesThe article proposes a theoretical understanding of in-context learning (ICL) in large pretrained language models such as GPT. The authors explain language models as meta-optimizers and understand ICL as implicit fine-tuning. They theoretically figure out that Transformer attention has a dual form of gradient descent and provide empirical evidence supporting their understanding. The article also suggests introducing momentum into attention as a potential method to further improve the performance of language models

deep-learning-theory-paper-discussion's People

Contributors

amartya-roy avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.