Riley F. Edmunds

I am currently the founder of Stride Labs, a startup building financial applications on blockchains. Before Stride, I spent three years at Bridgewater Associates working on machine learning and macro investing. During that time, I also completed a fellowship in deep learning with Google Brain and Jane Street.

I received my undergraduate degree in Computer Science from UC Berkeley in Spring 2019. At Berkeley, I was a research assistant in Dawn Song's Lab (part of BAIR) and Alice Agogino's Lab. I also built and led the research team at Machine Learning at Berkeley, a student organization of 80 members that researches with campus and industry labs and consults with tech companies.

While at Berkeley, I contributed to research published at ICRA, IROS, DLSW, and helped write a chapter on Adversarial Machine Learning in the "Artificial Intelligence Safety and Security".

Email  /  GitHub  /  LinkedIn  /  Scholar

Co-wrote and taught class on signal propagation & random matrix theory in deep neural networks as a 2019 Jane Street Fellow. Depth First Learning is a collaboration between NYU, FAIR, DeepMind, and Google Brain. Learn more here.

One of my first research reports at Bridgewater. Analysis of the unprecedented speed and magnitude of policy response to the COVID-19 crisis, comparing the timeline of asset reflation across the Great Depression, Global Financial Crisis, and current pandemic.

Work demonstrating that Tensegrity robots can learn locomotion policies even with severely limited sensory inputs using Mirror Descent Guided Policy Search (MDGPS).

Chapter surveying the field of Adversarial Machine Learning with connections to AI Safety.

Work desmonstrating that in a Meta-Learning context, adversarial attacks transfer between networks trained on different classification tasks drawn from Omniglot and mini-ImageNet.

A semi-supervised autoencoder model for fraud detection in event-streams having unbalanced class sizes and concept drift. Filed patent application based on invention.

The first tensegrity robot to achieve reliable locomotion on inclined surfaces of up to 24 degrees, with two novel multi-cable actuation policies, suited for steep incline climbing and speed, respectively.

For two years, built and led Machine Learning at Berkeley's research arm. We published 15+ papers at NeurIPS, ICML, ICLR, and ICCV. We hosted talks, poster sessions, with speakers including Ian Goodfellow and Andrej Karpathy. Developed working relationships with academic partners (BAIR, ICSI) and industry research collaborators (Google Brain, OpenAI).

We derived and implemented novel complex-valued layer functions for Convolutional Neural Networks, testing performance on frequency-domain problems: representation learning on MRI data and classification of SAR (Radar) data.

Work towards learning empirical transferability metrics between reinforcement learning tasks using autoencoder reconstruction errors and Model-Agnostic Meta-Learning.

A blog post on Neural Architecture Search's applications in Deep Learning, with an emphasis on crafting Domain Specific Languages to encode expressive yet tractable search spaces.

A tutorial on Bayesian Optimization and its applications in Deep Learning, with a focus on providing as much intuition behind the underlying theory as possible.

Using the Unity game engine and the HTC Vive, we created a 3D modeled virtual version of the UC Berkeley Campanile along with an interactive carillon.

Capsule connects the world through pictures. Travel to real world locations, pick up a capsule, and view images of everywhere else it's been. When you're done, travel somewhere else, take a picture, and leave it there for future explorers. Run up your score with frequent contributions, or use those points to add new capsules to the world. Happy travels!