Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Posts
Future Blog Post
Published:
This post will show up by default. To disable scheduling of future posts, edit config.yml and set future: false.
Building a simple code editor in a web browser
Published:
Building a code editor for a browser or web client.
Critiquing code via AI
Published:
How to generate code reviews with the suggested changes API automatically.
Notes on RepoFormer paper
Published:
Repoformer paper
Umbrela paper notes
Published:
Umbrela paper
Large Language Models can Accurately Predict Searcher Preferencess
Published:
Search Preference LLM labeling paper
Private python packages in Google cloud artifact registry
Published:
Google cloud artifact registry python packaging
On subsets and orderings
Published:
If you’re using a recent version of SciPY you might be using this idea already and not even know.
portfolio
Portfolio item number 1
Short description of portfolio item number 1
Portfolio item number 2
Short description of portfolio item number 2
publications
Nearest-neighbor matchup effects: accounting for team matchups for predicting March Madness
Published in Journal of Quantitative Analysis in Sports, 2015
We introduce our novel nearest-neighbor matchup effects framework, which presents a flexible way to account for team characteristics above and beyond team strength that may influence game outcomes. The paper was covered by several media outlets, click into the page to see one example.
Recommended citation: Hoegh, A., Carzolio, M., Crandell, I., Hu, X., Roberts, L., Song, Y., & Leman, S. C. (2015). Nearest-neighbor matchup effects: accounting for team matchups for predicting March Madness. Journal of Quantitative Analysis in Sports, 11(1), 29-37.
Download Paper | Download Slides
Smart Vision-Language Reasoners
Published in ICML 2024 & NeurIPS 2024, 2024
Images in Math AI Considered harmful? Not quite. This paper demonstrates improved performance on question answering problems in math by customizing the neural network architecture to pool information from both vision and text backbones. The improvements come from a custom QF layer which includes multihead self attention layers as well as a cross attention layer (vision & text). We fine tune the model using the smart-101 dataset presented in CVPR 2023.
Recommended citation: Roberts D, Roberts L. Smart Vision-Language Reasoners. arXiv preprint arXiv:2407.04212. 2024 Jul 5.
Download Paper
Algorithmic Differentiation of QR decompositions
Published in AlgoDiff 2024, 2024
Matrix Derivatives to propagate gradients through all QR matrices. This paper proves two formula mathematically equivalent for real matrices with real valued entries and non-equivalence for complex valued matrices. For complex field entries, we derivate an adjustment that enables library developers and maintainers to easily add support for complex matrices. NOTE: Linked paper is a preprint, full paper is in press at Springer.
Recommended citation: Roberts D, Roberts L. QR and LQ Decomposition Matrix Backpropagation Algorithms for Square, Wide, and Deep - Real or Complex - Matrices and Their Software Implementation. arXiv preprint arXiv:2009.10071. 2024
Download Paper
talks
Conference Proceeding talk 3 on Relevant Topic in Your Field
Published:
This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.
Talk on algorithmic differentiation of QR factorization
Published:
This is a talk at the Algorithmic Differentiation 2024 conference.
teaching
Teaching experience 1
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.