{"id":6,"date":"2026-02-12T20:59:11","date_gmt":"2026-02-12T20:59:11","guid":{"rendered":"https:\/\/armxss.com\/slt2026\/colibri-wp\/"},"modified":"2026-06-01T19:39:09","modified_gmt":"2026-06-01T19:39:09","slug":"colibri-wp","status":"publish","type":"page","link":"https:\/\/armxss.com\/slt2026\/","title":{"rendered":"Main"},"content":{"rendered":"
\n \n
\n \n \n
\n \n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n \n
\n

The\n Faculty of Mathematics and Mechanics<\/a> of the\n Yerevan State University<\/a>, in collaboration with\n ENSAE<\/a>, Paris, is organizing a<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Summer School on\n
Statistics and Learning Theory<\/h2>\n <\/div>\n <\/div>\n
\n \n \n
\n

which will be held on July 12 – 19, 2026<\/strong>, Armenia.<\/strong><\/p>\n <\/div>\n <\/div>\n

\n \n \n
\n

The target audience of the Summer School are university undergraduate seniors, Master and PhD students and researchers, as well as industry people interested in Probability, Statistics, Machine Learning and applications. Lectures will\n be in English.<\/p>\n <\/div>\n <\/div>\n

\n \n \n
\n

The above photos are from the previous event organized in Armenia, the \n Summer School on Cryptography, Statistics and Machine Learning<\/a>, June 29 \u2013 July 06, 2025.<\/p>\n <\/div>\n <\/div>\n

\n \n \n
\n

The participation fee is 450 EUR for academia participants and 600 EUR for industry participants. The participation fee will cover all local expenses, including registration fee, accommodation and full board. We have a special discounted\n participation fee of 250 EUR for undergraduate, master and PhD students from Armenian Universities. <\/strong> We intend to have some scholarships for academia participants to cover (fully or partially) their participation\n fees. For more details about how to apply for the scholarship, please see the\n Registration<\/a> page.<\/p>\n <\/div>\n <\/div>\n

\n \n \n
\n

Since the number of participants is very limited, the selection process is assumed for the registered participants. After receiving the participation approval, and only in that case<\/strong>, the payment should be made to the\n account provided in the confirmation email.<\/p>\n <\/div>\n <\/div>\n

\n \n \n
\n

The deadline for registration is June 15, 2026<\/strong>. Please find the registration form\n below<\/a>.<\/p>\n <\/div>\n <\/div>\n

\n \n \n
\n

The Arrival Day is July 12,<\/strong> 2026, and the Departure Day is July 19<\/strong>, 2026.<\/p>\n <\/div>\n <\/div>\n

\n \n \n
\n

For contact addresses, please visit the\n Contacts Page<\/a>.<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n \n
\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Lecturers<\/h2>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Michael Jordan<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

University of California, Berkeley, USA<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Eric Moulines<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

MBZUAI, UAE and EPITA, France<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Csaba Szepesv\u00e1ri<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

Google DeepMind and University of Alberta, Canada<\/p>\n <\/div>\n <\/div>\n

\n \n \n
\n

Cancelled<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n \n

Gregory Chirikjian<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

MBZUAI, UAE and University of Delaware, USA<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n \n

Pierre Alquier<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

ESSEC Business School, Singapore<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n \n

Yasin Abbasi-Yadkori<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

Sapient Intelligence, UK<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n \n

Avetik Karagulyan<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

CNRS\/L2S, France<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n \n
\n \n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Program<\/h2>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Yasin Abbasi-Yadkori: Uncertainty quantification in AI models<\/em><\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

Abstract: Language models can suffer from hallucinations, producing responses with low factual accuracy. Since these models fundamentally represent probability distributions over text, questions of truthfulness are naturally connected\n to the concept of statistical uncertainty. In this lecture, I will discuss uncertainty estimation for AI models, beginning with high-probability confidence sets in simple linear models and progressing to methods for quantifying uncertainty\n in contemporary large language models.<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Pierre Alquier: An Introduction to PAC-Bayesian Theory and Its Modern Applications<\/em><\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

Abstract: The PAC-Bayesian theory provides tools to understand the accuracy of Bayes-inspired algorithms that learn probability distributions on parameters. This theory was initially developed by McAllester about 20 years ago, and\n applied successfully to various machine learning algorithms in various problems. Recently, it led to tight generalization bounds for deep neural networks, a task that could not be achieved by standard “worst-case” generalization\n bounds such as Vapnik-Chervonenkis bounds. The first lecture is a brief introduction to PAC-Bayes bounds and explain the core ideas and the main applications. I will also provide detailed proofs of some of the core results. In the\n second lecture, I will discuss more theoretical aspects. In particular, I will highlight the application of PAC-Bayes bounds to derive minimax-optimal rates of convergence in classification and in regression, and the connection to\n mutual-information bounds.<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Gregory Chirikjian: Stochastic Models and Lie Groups in Robotics<\/em><\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

Abstract:<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Michael Jordan: Machine Learning as a Blend of Computational, Economic, and Statistical Perspectives <\/em><\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

Abstract:<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Avetik Karagulyan: Langevin Sampling and Stochastic Optimization<\/em><\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

Abstract: The lectures explore the interplay between sampling and optimization, presenting sampling from complex distributions as an optimization problem. We introduce Langevin dynamics, a method that extends gradient descent\n with stochastic perturbations to enable efficient exploration of target distributions. By drawing parallels with stochastic gradient descent (SGD), we demonstrate how optimization principles can be adapted for probabilistic sampling.\n <\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Eric Moulines: TBA<\/em><\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

Abstract: TBA<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n \n
\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Organizers<\/h2>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Scientific Committee<\/h2>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Arnak Dalalyan<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

ENSAE Paris\/ CREST, Institut Polytechnique de Paris<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Karen Keryan<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

YSU, AUA<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Michael Poghosyan <\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

YSU, AUA<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Organizing Committee<\/h2>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n \n

Julieta Boyajyan<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

YSU, AUA<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n \n

Karen Keryan<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

YSU, AUA<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n
\n \n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/a>\n <\/div>\n <\/div>\n
\n \n
\n \n \n

Michael Poghosyan<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n

YSU, AUA<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n \n
\n \n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Registration<\/h2>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n \n
\n

To register for participation, please complete the form using the link provided below. Please note that the number of participants is limited; therefore, a selection process will be applied to all registered applicants. The Organizing\n Committee will review all applications and notify applicants whether their participation has been approved or declined. Payment should be made only after<\/strong> receiving an official confirmation email and only in the case of approval<\/strong>,\n using the account details provided therein.<\/p>\n

The participation fee is 450 EUR<\/strong> for academic participants and 600 EUR<\/strong> for industry participants. The fee covers all local expenses, including the registration fee, accommodation, and full board. A\n special discounted participation fee of 250 EUR<\/strong> is available for undergraduate, master\u2019s, and PhD students from Armenian universities. In addition, a limited number of scholarships will be available for academic\n participants to fully or partially cover the participation fee. To apply for a scholarship, please select the corresponding checkbox in the registration form; shortlisted applicants will be contacted for further details.<\/p>\n

The registration deadline is June 15, 2026<\/strong>.<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
Link to the Registration Form<\/span><\/a>\n <\/span>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n \n
\n \n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Schedule<\/h2>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n \n
\n

The schedule will be ready in June<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n \n
\n \n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Sponsors<\/h2>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

Organizations<\/h4>\n <\/div>\n <\/div>\n
\n \n \n
\n