--- title: "Chris J. Maddison — author profile and papers" description: "Chris J. Maddison, University of Toronto: 2 papers on Elamur Research (2016–2024), h-index 15 — affiliations, topics, co-authors and who cites them in…" canonical: "https://research.elamur.ai/free/en/authors/chris-j-maddison/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/chris-j-maddison/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Chris J. Maddison # Chris J. Maddison Chris J. Maddison works in machine learning, and the papers here span two quite different questions. One is how to train models that contain discrete random choices, where the usual gradient machinery breaks down. The other is how to forecast what large language models will be able to do without paying to train them. Both are approached through probabilistic modelling and careful statistical analysis, whether of distributions on the probability simplex or of benchmark results gathered across many existing systems. The most cited of Maddison's papers here is [The Concrete Distribution](https://md.research.elamur.ai/free/en/papers/maddison-2016-concrete-distribution-continuous-relaxation-discrete/) from 2016, written with Mnih and [Yee Whye Teh](https://md.research.elamur.ai/free/en/authors/yee-whye-teh/). It starts from the fact that the reparameterization trick lets gradients flow through continuous stochastic nodes but not discrete ones, which usually forces a fallback to score-function estimators. Those estimators are unbiased but tend to be noisy. The paper's answer is a family of continuous distributions on the simplex, with densities in closed form, that stand in for categorical variables during optimisation. The more recent [Observational Scaling Laws](https://md.research.elamur.ai/free/en/papers/ruan-2024-observational-scaling-laws-predictability-language/) from 2024 turns to [language models and pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) and to [evaluation and benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/). Rather than training a ladder of models of increasing size, it draws on around a hundred publicly available models and their leaderboard scores. It then uses a principal-components decomposition to recover a small set of general capability measures that scale with the logarithm of compute within each model family. Maddison is currently at the [University of Toronto](https://md.research.elamur.ai/free/en/universities/university-of-toronto/) and the [Vector Institute](https://md.research.elamur.ai/free/en/labs/vector-institute/). Earlier bylines include the [University of British Columbia](https://md.research.elamur.ai/free/en/universities/university-of-british-columbia/), [Google DeepMind](https://md.research.elamur.ai/free/en/companies/google-deepmind/), [Google](https://md.research.elamur.ai/free/en/companies/google/) and the [University of Oxford](https://md.research.elamur.ai/free/en/universities/university-of-oxford/). Co-authors on the papers here include Yee Whye Teh and [Tatsunori Hashimoto](https://md.research.elamur.ai/free/en/authors/tatsunori-hashimoto/). Chris J. Maddison has 2 papers in our library, in 2016–2024, most of them on [Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/), [Evaluation & benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/); often with [Tatsunori Hashimoto](https://md.research.elamur.ai/free/en/authors/tatsunori-hashimoto/), [Yee Whye Teh](https://md.research.elamur.ai/free/en/authors/yee-whye-teh/). 2 papers in our library · 2016–2024 [OpenAlex](https://openalex.org/A5054711904) · [ORCID](https://orcid.org/0009-0007-9796-7232) · [**h-index 15**](https://openalex.org/authors/A5054711904) · [85 works](https://openalex.org/works?filter=author.id:A5054711904) · [18,031 citations](https://openalex.org/works?filter=author.id:A5054711904&sort=cited_by_count:desc) OpenAlex, 2026-10-14 [📄 2 works in our library](https://md.research.elamur.ai/free/en/authors/chris-j-maddison/#papers) · [🔗 312 citations in our library](https://md.research.elamur.ai/free/en/authors/chris-j-maddison/cited-by/) · [1 more in our library without a page yet](#also) · [⭐ Paper Star rank #1437](https://md.research.elamur.ai/free/en/paper-stars/page/15/#star-1437) The profile is the library's match of the name; two people with one name are not always told apart. Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Chris+J.+Maddison) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Chris+J.+Maddison) · [dblp](https://dblp.org/search?q=Chris+J.+Maddison) · [ACL Anthology](https://aclanthology.org/search/?q=Chris+J.+Maddison) ## Affiliations [🎓**University of Toronto** Toronto · Canada 2014–2026 · 38 papers*current*](https://md.research.elamur.ai/free/en/universities/university-of-toronto/) · [🔬**Vector Institute** Toronto · Canada 2020–2026 · 23 papers*current*](https://md.research.elamur.ai/free/en/labs/vector-institute/) · [🏢**Google** Mountain View · United States 2017–2019 · 7 papers](https://md.research.elamur.ai/free/en/companies/google/) · [🏢**Google DeepMind** London · United Kingdom 2016–2019 · 4 papers](https://md.research.elamur.ai/free/en/companies/google-deepmind/) · [🎓**University of Oxford** Oxford · United Kingdom 2017–2019 · 4 papers](https://md.research.elamur.ai/free/en/universities/university-of-oxford/) · [🎓**University of British Columbia** Vancouver · Canada 2012–2021 · 3 papers](https://md.research.elamur.ai/free/en/universities/university-of-british-columbia/) From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 2 more institutions on single bylines are not listed. ## Topics [Language models & pretraining (1)](https://md.research.elamur.ai/free/en/topics/language-models/) · [Evaluation & benchmarks (1)](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) ## Works by year ### 2020s [**2024** 1 paper](https://md.research.elamur.ai/free/en/papers/ruan-2024-observational-scaling-laws-predictability-language/) ### 2010s [**2016** 1 paper](https://md.research.elamur.ai/free/en/papers/maddison-2016-concrete-distribution-continuous-relaxation-discrete/) ## Papers The most cited first (OpenAlex and our library). 1. **[The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables](https://md.research.elamur.ai/free/en/papers/maddison-2016-concrete-distribution-continuous-relaxation-discrete/)** *(2016 · 1,569 citations)* The reparameterization trick lets gradients pass through continuous random nodes in stochastic computation graphs. 2. **[Observational Scaling Laws and the Predictability of Language Model Performance](https://md.research.elamur.ai/free/en/papers/ruan-2024-observational-scaling-laws-predictability-language/)** *(2024 · 28 citations)* Classical scaling laws relate the training compute budget to the quality of a language model. Building them requires training a series of models of different sizes oneself, which few can afford. The authors propose an "observational" approach: instead of training, they take about a hundred open models and their… ## Also in our library 1 more work of theirs is in our library without a page here yet; it will appear as we publish them. ## Co-authors here 2 co-authors with a page, the most papers together first. - [Tatsunori Hashimoto](https://md.research.elamur.ai/free/en/authors/tatsunori-hashimoto/) - [Yee Whye Teh](https://md.research.elamur.ai/free/en/authors/yee-whye-teh/) ## Cited by Papers of our library cite their works 312 times (of the 140,884 papers whose bibliographies we read); [8 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/chris-j-maddison/cited-by/).