The Criteo AI Lab is pioneering innovations in computational advertising. As the centre of scientific excellence in the company, we deliver both fundamental and applied scientific leadership through published research, product innovations and new technologies powering the companyâ€™s products.
The Criteo AI Lab operates within the spectrum of two main roles:
Applied Research:Â Our Scientists fully leverage the advantage of working in a machine-learning driven organization by partnering closely with our product and engineering counterparts to deliver cutting-edge solutions to the challenges in online advertising.
Academic Contributions:Â The Research Scientists at Criteo are encouraged and fully supported to publish their works at international conferences, collaborate with academic partners, file for patents, release datasets and help establish the state-of-art in computational advertising.
- Online learning, bandit algorithms: having to process the terabytes of data that Criteo collects every day and trying to get the most out of it requires having at hand ML algorithms capable of adapting in effective and relevant way. Building online learning algorithms and the accompanying theory (cf. our papers on bandits algorithms) is therefore at the heart of our research.
- Learning in repeated mechanism / markets: a topic of crucial importance concerns the interactions between a few AI-based systems the potential problems that can arise, which is is particularly important for B2B companies having automated interactions and thus facing the problem of learning from repeated interactions. A mechanism â€“ e.g. an auction (advertising), a contract, a market (electricity, goods, 5G...) -- rules these interactions, and brings the question on how AI agents can learn from each other, whether the whole systems stabilizes, how wealth is distributed across participants.
- Causality learning: the problem of identifying causal mechanisms is an integral part of the scientific method and has been traditionally posed in varied disciplines such as earth, life and social sciences. There is a lot of involvement of the research group on an increasing ladder of difficulty regarding causality: estimate the effect of interventions (e.g. how to design a protocol to measure the causal effect of a system up to a desired significance level), predict the effect of proposed interventions (e.g. how the causal effect of a system is changed if we vary some intensity parameter), optimize the effect of interventions (e.g. what is the ideal intensity parameter of a system to maximize its causal effect under some feasibility constraints)
- Natural language processing: a huge effort is put on having a finer and finer understanding of contextual information collected on publisher websites, improving search capabilities for some of our products (e.g. Retail Media, Universal Catalog), and being able to handle multilingual data requires us to be equipped with the most advanced NLP assets pertaining to our needs.
- Research on recommendation and learning from aggregated data: in the goal of improving our recommender systems across Criteo products using cutting-edge advancements, we explore new techniques based on deep learning, counterfactual risk minimisation, and Bayesian machine learning, with a special focus on learning from aggregated data.
- Privacy-preserving learning and fairness: the topic of learning privacy-preserving models that, in addition, preserve fairness, has become central to our activities.
- Computer vision: there is a line of research developed around deep learning (and in particular generative DL
What you'll do
- Identify new research opportunities at Criteo and lead the exploration of these ideas and pursue patents/publications where appropriate.
- Interact with other teams to define interfaces and understand and resolve dependencies
- Maintain world-class academic credentials through publications, presentations, external collaborations and service to the research community.
- Understand and shape the product direction by contributing innovative ideas, specifically as a result of data mining and experimental data modelling
- Develop high-performance algorithms, test and implement the algorithms in scalable and product-ready code.
- Mentor team members, oversee the creation of technical documents and work towards establishing Criteo as a centre of excellence in computational advertising.
Who you are
- PhD in Machine Learning or a related field along with one to three years of experience.
- Implementation experience with languages such as Python, Perl, Ruby, Java, C#, Scala etc.
- Excellent track record in conducting and reporting results of original and collaborative research with publications.
- Hands-on skills in sourcing, cleaning, manipulating and analyzing large volumes of data.