Bradley Brown

Hello! I am a Computer Science Master's student at the University of Oxford, supervised by Professor Ronald Clark and supported by the Clarendon Fund. I am also working as a visiting researcher in the Scaling Intelligence Lab at Stanford with Professor Azalia Mirhoseini.

I obtained my undergraduate degree at the University of Waterloo studying Software Engineering with a joint major in Combinatorics and Optimization. Previously, I was a Research Scientist intern at NVIDIA’s Toronto AI lab, Layer 6 AI, and Akasha Imaging.

[ Email / Github / Twitter / Google Scholar / LinkedIn ]

News

June 2024: Starting as a Visiting Student Researcher in the Scaling Intelligence Lab at Stanford with Professor Azalia Mirhoseini!
April 2023: Paper @ CVPR 2023 - Our work creating a generative model of open-world 3D scenes, NF-LDM, was accepted to CVPR 2023!
April 2023: Tiny Paper @ ICLR 2023 - Our work examing a limitation of increased scale in LLMs was accepted to the ICLR 2023 tiny paper track.
January 2023: Paper @ ICLR 2023 - Our work Verifying the Union of Manifolds Hypothesis for Image Data was accepted to ICLR 2023!
October 2022: Workshop Paper @ NeurIPS 2022 - Our work Relating Regularization and Generalization through the Intrinsic Dimension of Activations was accepted to OPT 2022 and HITY 2022.
September 2022: Internship - Returning to NVIDIA as a Research Scientist intern in the Toronto AI lab
September 2022: RecSys Competion - Our team placed 2 / 303 teams in the 2022 RecSys Challenge, read about our approach here!
February 2022: Internship - Joined Layer 6 AI as a Machine Learning Researcher in their generative modelling team.
May 2021: Internship - Joined NVIDIA as a Research Scientist intern in the Toronto AI lab.
July 2021: Paper @ ICCV 2021 - Our work Towards Rotation Invariance in Object Detection was accepted at ICCV 2021.
July 2020: Internship - Joined Akasha Imaging (now part of Intrinsic) as a Computer Vision and Robotics researcher.

Research

	CodeMonkeys: Scaling Test-Time Compute for Software Engineering Ryan Ehrlich^, Bradley Brown^, Jordan Juravsky^*, Ronald Clark, Christopher Ré, Azalia Mirhoseini [ Paper / Code / Trajectories / Blog Post (Monkey SWE, Monkey Do) ] Building a system for solving SWE-bench issues that is explicitly designed for scaling test-time compute.
	Large Language Monkeys: Scaling Inference Compute with Repeated Sampling Bradley Brown^, Jordan Juravsky^, Ryan Ehrlich^, Ronald Clark, Quoc V. Le, Christopher Ré, Azalia Mirhoseini Preprint*. [ Paper / Code / Dataset / Blog Post ] Demonstrating that increasing the amount of inference compute through repeated sampling leads to large improvements in coverage - the fraction of problems solved by any attempt - across a variety tasks, models, and sample budgets. This makes it possible, and sometimes cost-effective, to amplify weaker models with many samples and outperform single attempts from more capable models.
	Hydragen: High-Throughput LLM Inference with Shared Prefixes Jordan Juravsky^, Bradley Brown^, Ryan Ehrlich^, Daniel Y. Fu, Christopher Ré, Azalia Mirhoseini Preprint*. [ Paper / Code ] Introducing an exact, simple (no custom CUDA) implementation of attention that can accelerate LLM throughput by over 30x for problems containing shared prefixes and large batch sizes.
	NeuralField-LDM: Scene Generation with Hierarchical Latent Diffusion Models Seung Wook Kim^, Bradley Brown^, Kangxue Yin, Karsten Kreis, Katja Schwarz, Daiqing Li, Robin Rombach, Antonio Torralba, Sanja Fidler IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2023. [ Project Page / Paper ] Building a generative model of open-world 3D scenes trained on real-world in-the-wild data.
	Verifying the Union of Manifolds Hypothesis for Image Data Bradley C.A. Brown, Anthony L. Caterini, Brendan Leigh Ross, Jesse C. Cresswell, Gabriel Loaiza-Ganem International Conference on Learning Representations (ICLR) 2023. [Paper / Video / Code] Extending the manifold hypothesis to support natural image data lying on a union of manifolds with varying intrinsic dimension. Show increased performance in generative modelling and image classification tasks by designing models with an inductive bias for this structure.
	Language Models Inversely Scale on Piecewise Function Evaluation with Biased Examples Jordan Juravsky^, Bradley Brown^, Atif Mahmud^, Ryan Ehrlich^, Wais Shahbaz^* Tiny Paper at the International Conference on Learning Representations (ICLR) 2023. [Paper] Demonstrating that large language models (LLMs) can be misled by providing them with factually correct, but unrepresentative/biased examples, in the context of integer-to-integer piecewise functions.
	Relating Regularization and Generalization through the Intrinsic Dimension of Activations Bradley C.A. Brown, Jordan Juravsky, Anthony L. Caterini, Gabriel Loaiza-Ganem NeurIPS 2022 workshops: OPT 2022 and HITY 2022. [ Paper / Code ] Investigating how the intrinsic dimension of activations in deep neural networks are affected by regularization, correlated with improved validation performance and are coupled with the effects of sudden generalization (grokking).
	Session-based Recommendation with Transformers Yichao Lu, Zhaolin Gao, Zhaoyue Cheng, Jianing Sun, Bradley Brown, Guangwei Yu, Anson Wong, Felipe Pérez, Maksims Volkovs Proceedings of the Recommender Systems Challenge 2022. [Paper] Leveraging transformers and self-supervised learning techniques to achieve 2/300+ teams on the RecSys session-based recommendation system challenge.
	Towards Rotation Invariance in Object Detection Agastya Kalra, Guy Stoppi, Bradley Brown, Rishav Agarwal, Achuta Kadambi International Conference on Computer Vision (ICCV) 2021. [ Paper / Video / Code ] Proposing a mathematically sound rotation augmentation scheme and loss modification for object detection models that leads to better rotation invariance/equivariance.

Template from this website, adapted from this website.