A practical look at how curiosity, good mentoring, and consistent feedback draw students into meaningful research work.
- BIO
Pliff Jenkins
"I have already bought and tested many, many templates. But this Canvas HTML Template is really the best and most comprehensive for me. I highly recommend it."
Academic Positions
Professor
MIT, Course - Jun 3rd 2022
Assistant Professor
MIT, Course - Sep 11th, 2017
Assistant Professor
St. John’s University, Kishanattam, Kerala - Mar 7th, 2011
Visiting Ass. Professor
ADR-Centric Juridical University - Dec 8th, 2010
Education & Training
Ph.D. in Course
University of NY - 2021
B. Sc. in Applied Course
University of NY - 2017
B. Sc. in Statistics
Institute for Mathametics - 2015
Rewards
Distinguished Fellow of the Royal Statistical Society
2023
Awarded for sustained contributions to statistical learning theory and the advancement of optimal transport methods. The fellowship recognises a decade of peer-reviewed research, the supervision of doctoral candidates, and a commitment to making advanced quantitative methods accessible to students and practitioners across applied fields and disciplines worldwide.
Camvat Education Reward
2020
Presented by the Camvat Foundation for outstanding teaching and curriculum design in higher education. The award honours faculty who reshape how core quantitative subjects are taught, blending rigorous theory with practical, hands-on learning.
Listen & lecture Statistical
2020
Watch video here.
This recorded lecture explores how statistical reasoning shapes modern decision-making, drawing on case studies from public policy and applied research to show why sound inference matters in practice for everyday analysts everywhere.
This recorded lecture explores how statistical reasoning shapes modern decision-making, drawing on case studies from public policy and applied research to show why sound inference matters in practice for everyday analysts everywhere.
Year of the Knowledge Rewards
2018
Watch video here.
Honoured for a body of work that brought rigorous methods to a wider audience, this recognition celebrates a year of seminars, open lectures, and mentoring aimed at building lasting quantitative literacy among emerging researchers everywhere.
Honoured for a body of work that brought rigorous methods to a wider audience, this recognition celebrates a year of seminars, open lectures, and mentoring aimed at building lasting quantitative literacy among emerging researchers everywhere.
Works
Education of One.
Published on: 05th Jan, 2022
A concise study on how individualised instruction reshapes learning outcomes, drawing on classroom data and longitudinal surveys. The book argues that tailored feedback, when paired with measurable goals, consistently outperforms one-size-fits-all teaching across diverse subjects.
The Psycolodgy of Knowledge.
Published on: 11th Oct, 2021
An exploration of how the mind acquires, retains, and applies new knowledge over time. Combining cognitive science with practical study techniques, this work offers readers a clear framework for learning faster and remembering far more of what they read.
Experience
Dec 2022 ─ Present
Leading the applied statistics group, supervising doctoral research and developing graduate coursework in modern optimisation methods.
Mar 2021 ─ Dec 2022
Conducted postdoctoral work on optimal transport, publishing several peer-reviewed papers and co-teaching an advanced seminar series.
Feb 2019 ─ Mar 2021
Served as visiting lecturer, designing introductory courses in probability and statistical inference.
Jan 2017 ─ Feb 2019
Completed doctoral research in statistics and began mentoring undergraduate students in quantitative methods.
Math Expert
Deep command of probability, optimisation, and the proofs behind modern statistics.
Longtime Experience
More than a decade of teaching, research, and supervision across leading universities.
Loyalty
A steadfast commitment to students, collaborators, and the institutions I serve.
Hard Worker
Disciplined, detail-driven work ethic applied to long and demanding research.
Great Leadership
Guiding research teams with clarity, fairness, and a focus on shared, lasting goals.
Good Speakers
Engaging, accessible presentations that turn complex ideas into clear, memorable talks.
Journal
A Mathematical Model of Self-Attention Dynamics
Borjan Geshkovski, Cyril Letrouit, Yury Polyanskiy, and Philippe Rigollet (2023)
Preprint
abcdefg: 123456
Abstract
We study the dynamics of self-attention as an interacting particle system and show that tokens evolve toward a small number of clusters as depth increases. Using tools from optimal transport, we characterise the limiting configuration and prove quantitative convergence rates that match the behaviour observed empirically in large trained transformer architectures.
On the Emergence of Clusters in Transformer Models
Borjan Geshkovski, Cyril Letrouit, Yury Polyanskiy, and Philippe Rigollet (2023)
Closed
abcdefg: 123456
Abstract
We examine how token representations group together as information propagates through successive attention layers. Casting the process as a measure-valued flow, we identify the conditions under which distinct clusters form and quantify how their number depends on temperature, depth, and the spectral properties of the underlying weight matrices used.
Mean-Field Limits for Deep Residual Networks
Borjan Geshkovski, Cyril Letrouit, Yury Polyanskiy, and Philippe Rigollet (2023)
Info
abcdefg: 123456
Abstract
We derive a mean-field description of deep residual networks in which the discrete layers are replaced by a continuous-time flow on the space of probability measures. This viewpoint clarifies how skip connections stabilise training and yields rigorous bounds on the gap between the finite-depth network and its idealised limiting dynamics here.
Convergence Rates in Optimal Transport Estimation
Borjan Geshkovski, Cyril Letrouit, Yury Polyanskiy, and Philippe Rigollet (2023)
Preprint
abcdefg: 123456
Abstract
We establish finite-sample convergence rates for estimators of optimal transport maps between high-dimensional distributions. Combining minimax analysis with smoothness assumptions on the underlying densities, we obtain bounds that improve upon prior work and clarify the precise role played by dimension and regularity in the estimation problem.Research
Research Summary
My research blends statistical learning theory with optimal transport methods.
I study how modern machine learning models organise information, developing rigorous tools to explain their behaviour and to guide the design of more efficient, reliable, and interpretable learning systems.
Interests
Learning
Minimax optimality
Optimal transport
Optimization
Algorithms and complexity
Latest Blogs
Reflections on the lessons that shaped my teaching, gathered from years of working closely with curious and ambitious students alike.
A candid guide to online education: what genuinely works, what falls short, and how learners can get the most from remote study today.
An honest overview of open research positions, what each role involves, and the qualities our group looks for in new applicants now.
Contact
My Office
795 Folsom Ave, Suite 600San Francisco, CA 94107
Phone: (1) 8547 632521
Fax: (1) 11 4752 1433
Email: info@canvas.com