PDF e-Pub

## Section: Partnerships and Cooperations

### International Initiatives

#### Inria International Labs

##### 6PAC (IIL CWI-Inria)
• Scientific leaders: Benjamin Guedj, Peter Grünwald.

• Other members: Emilie Kaufmann (Inria LNE, EPI SequeL), Wouter Koolen (CWI).

• Title: Making Probably Approximately Correct Learning Active, Sequential, Structure-aware, Efficient, Ideal and Safe

• International Partner (Institution - Laboratory - Researcher):

• CWI (Netherlands) - Machine Learning Group - Peter Grünwald (head)

• Start year: 2018, renewed for 2019 and 2020

• This project roots in statistical learning theory, which can be viewed as the theoretical foundations of machine learning. The most common framework is a setup in which one is given $n$ training examples, and the goal is to build a predictor that would be efficient on new (similar) data. This efficiency should be supported by PAC (Probably Approximately Correct) guarantees, e.g. upper bounds on the excess risk of a predictor that hold with high probability. Such guarantees however often hold under stringent assumptions which are typically never met in real-life application, e.g., independent, identically distributed data. More realistic modelling of data has triggered many research efforts in several directions: first, accommodating possible data (e.g., dependent, heavy-tailed), and second, in the direction of sequential learning, in which the predictor can be built on the fly, while new data is gathered. We believe that an ever more realistic paradigm is active learning, a setup in which the learner actively requests data (possibly facing constraints, such as storage, velocity, cost, etc.) and adapts its queries to optimize its performance. The 3-years objective of 6PAC (where 6 stands for Sequential, Active, Efficient, Structured, Ideal, Safe - the six research directions we intend to contribute to) is to pave the way to new PAC generalization and sample-complexity upper and lower bounds beyond batch learning. Our ambition is to contribute to several learning setups, ranging from sequential learning (where data streams are collected) to adaptive and active learning (where data streams are requested by the learning algorithm).

#### Inria International Partners

##### Declared Inria International Partners
• A byproduct of Benjamin Guedj's sabbatical position at University College London (UCL) since Dec 2018 is a strengthened link between UCL and Inria. DGDS has established contact with UCL President in April 2019 and a MoU has been signed between UCL and Inria in December 2019. A research group (known as Inria@UCL) has been established by Benjamin Guedj within UCL, Department for Computer Science, Centre for Artificial Intelligence. Inria@UCL initiative is expected to grow in 2020 and possibly evolve into a joint team or more. A strategic partnership between Inria and UCL will be explored in 2020.

• SIMERGE

• Title: Statistical Inference for the Management of Extreme Risks and Global Epidemiology

• International Partner (Institution - Laboratory - Researcher):

• UGB (Senegal) - LERSTAD - Abdou Ka Diongue

• Serge Iovleff and Sophie Dabo-Niang are associated members of SIMERGE.