Data Fest 2020

Math Optimization

Hi! Welcome to the optimization track on Datafest! Here we will discuss both theory and practice of modern optimization. We say really modern because the results and methods explained in the section are used either to conduct bleeding-edge research or train neural networks faster and larger.

In this section, You’ll know about:

  • Automatic Evaluation and Tuning System for Clustering Algorithms from Sergey Muravyov 
  • Structural Optimization for Composite Models with Evolutionary Algorithms from Nikolay Nikitin
  • NVIDIA Tricks for Training Large Neural Nets with AMP/TF32 from Denis Timonin
  • Riemannian Optimization from Maxim Kochurov
  • Riemannian Optimization for Quantum Technologies by Ilia Luchnikov and Alexander Ryzhov
  • Integrals in Extremely Large Dimensions from Yury Maximov
  • Polynomial Optimization from Olga Kuryatnikov

Track program

Automatic evaluation and tuning system for clustering algorithms

Automatic evaluation and tuning system for clustering algorithms

Sergey Muravyov

Structural Learning for Composite Models Using Evolutionary Optimisation

Structural Learning for Composite Models Using Evolutionary Optimisation

Nikolay Nikitin

Fast training with AMP/TF32 using TensorCores on NVIDIA GPU

Fast training with AMP/TF32 using TensorCores on NVIDIA GPU

Denis Timonin

Riemannian optimization

Riemannian optimization

Maxim Kochurov

Riemannian optimization for quantum technologies

Riemannian optimization for quantum technologies

Ilia Luchnikov, Alexander Ryzhov

How to Compute an Integral in a High Dimension, and Why is it an Optimization Problem?

How to Compute an Integral in a High Dimension, and Why is it an Optimization Problem?

Yury Maximov

Polynomial optimization: introduction and applications in data analysis

Polynomial optimization: introduction and applications in data analysis

Olga Kuryatnikova

Cookies help us deliver our services. By using our services, you agree to our use of cookies.

Learn More