Open in app

Sign in

Write

Sign in

Jason McEwen
Jason McEwen

210 Followers

Home

About

Apr 19

Towards Generative Geometric AI

Generative AI for 360° panoramic images — Generative AI has made remarkable progress in recent years, enabling machines to generate images, text, and even music. However, a number of data modalities are still missing. Some of the most notable generative models include GPT-4, a language model that can generate human-like text, and DALL-E 2, an image generation…

Geometry

5 min read

Towards Generative Geometric AI
Towards Generative Geometric AI
Geometry

5 min read


Published in

Towards Data Science

·Mar 15

Hybrid Discrete-Continuous Geometric Deep Learning

Scalable and equivariant spherical CNNs by DISCO convolutions — No existing spherical convolutional neural network (CNN) framework is both computationally scalable and rotationally equivariant. Continuous approaches capture rotational equivariance but are often prohibitively computationally demanding. Discrete approaches offer more favorable computational performance but at the cost of equivariance. We develop a hybrid discrete-continuous (DISCO) group convolution that is simultaneously…

Deep Learning

7 min read

Hybrid Discrete-Continuous Geometric Deep Learning
Hybrid Discrete-Continuous Geometric Deep Learning
Deep Learning

7 min read


Published in

Towards Data Science

·Mar 6

Geometric Deep Learning on Groups

Continuous vs discrete approaches on the sphere — Ideally geometric deep learning techniques on groups would encode equivariance to group transformations, to provide well-behaved representation spaces and excellent performance, while also being computationally efficient. However, no single approach provides both of these desirable properties. Continuous approaches offer excellent equivariance but with a very large computational cost. Discrete approaches…

Geometric Deep Learning

6 min read

Geometric Deep Learning on Groups
Geometric Deep Learning on Groups
Geometric Deep Learning

6 min read


Published in

Towards Data Science

·Sep 28, 2022

Scaling Spherical Deep Learning to High-Resolution Input Data

Scattering networks on the sphere for scalable and rotationally equivariant spherical CNNs — Conventional spherical CNNs are not scalable to high resolution classification tasks. In this post we present spherical scattering layers — a novel spherical layer that reduces the dimensionality of the input data while retaining relevant information, while also being rotationally equivariant. Scattering networks work by employing predefined convolutional filters from…

Machine Learning

8 min read

Scaling Spherical Deep Learning to High-Resolution Input Data
Scaling Spherical Deep Learning to High-Resolution Input Data
Machine Learning

8 min read


Published in

Towards Data Science

·Jul 27, 2022

Democratizing Geometric AI for 360° Spherical Data

Unlocking AI for 360° spherical data — While AI is now commonplace for standard types of data, such as structured, sequential and image data, the application of AI is severly curtailed for other more complex forms of data. These more complex datasets typical exhibit non-trivial geometry. The field of geometric AI, or geometric deep learning, has emerged…

Geometric Deep Learning

4 min read

Democratizing Geometric AI for 360° Spherical Data
Democratizing Geometric AI for 360° Spherical Data
Geometric Deep Learning

4 min read


Published in

Towards Data Science

·Jul 25, 2022

A Brief Introduction to Geometric Deep Learning

AI for complex data — Deep learning is hard. While universal approximation theorems show that sufficiently complex neural networks can in principle approximate “anything”, there is no guarantee that we can find good models. Great progress in deep learning has nevertheless been made by judicious choice of model architectures. These model architectures encode inductive biases…

Thoughts And Theory

8 min read

A Brief Introduction to Geometric Deep Learning
A Brief Introduction to Geometric Deep Learning
Thoughts And Theory

8 min read


Published in

Towards Data Science

·May 16, 2022

Walkable 360° Video

Geometric AI allows you to step inside 360° VR photos and videos — Today’s virtual reality (VR) experiences fall short of the realism users expect. Current VR technology essentially falls into two categories: those based on 3D models with computer generated imagery (CGI); and those based on panoramic 360° imagery. CGI-based experiences are interactive and support novel views but are far from photorealistic…

Metaverse

3 min read

Walkable 360° Video
Walkable 360° Video
Metaverse

3 min read


Published in

Towards Data Science

·May 13, 2022

Learnt Harmonic Mean Estimator for Bayesian Model Selection

Machine learning assisted computation of the marginal likelihood — Bayesian model comparison provides a principled statistical framework for selecting an appropriate model to describe observational data, naturally trading off model complexity and goodness of fit. However, it requires computation of the Bayesian model evidence, also called the marginal likelihood, which is computationally challenging. …

Bayesian Statistics

9 min read

Learnt Harmonic Mean Estimator for Bayesian Model Selection
Learnt Harmonic Mean Estimator for Bayesian Model Selection
Bayesian Statistics

9 min read


Published in

Towards Data Science

·May 3, 2022

Efficient Generalized Spherical CNNs

Hybrid rotationally equivariant spherical CNNs — Notions of spherical convolution offer a promising route to unlocking the potential of deep learning for the variety of problems in which spherical data are prevalent. However, the introduction of non-linearity is a challenge. In this post we explore how ideas originating in quantum physics may be applied to overcome…

Machine Learning

10 min read

Efficient Generalized Spherical CNNs
Efficient Generalized Spherical CNNs
Machine Learning

10 min read


Published in

Towards Data Science

·Apr 26, 2022

Geometric Deep Learning for Spherical Data

Spherical CNNs By encoding an understanding of the translational symmetry of the physical world, convolutional neural networks (CNNs) have revolutionised computer vision. …

AI

9 min read

Geometric Deep Learning for Spherical Data
Geometric Deep Learning for Spherical Data
AI

9 min read

Jason McEwen

Jason McEwen

210 Followers

Professor of Astrostatistics, UCL | Founder & CEO, CopernicAI

Following
  • Emma Boudreau

    Emma Boudreau

  • Benjamin Hardy, PhD

    Benjamin Hardy, PhD

  • Michael Bronstein

    Michael Bronstein

  • Xichu Zhang

    Xichu Zhang

  • Theo McCaie

    Theo McCaie

See all (13)

Help

Status

About

Careers

Blog

Privacy

Terms

Text to speech

Teams