Hello! I am Tristan Karch, a Ph.D. candidate at Inria in the Flowers team under the supervision of Pierre-Yves Oudeyer and Clément Moulin-Frier. My research is mainly on Vygotskian Autotelic Artifical Agents: agents that use language to imagine their own goals, for planning, reasoning and learning about them. I also work on emergent communication in Reinforcement Learning settings and in Language Games.

Short Bio

I have a diverse academic background ranging from Mechanical Engineering to Machine Learning. I started by completing a BSc and MSc at Ecole Polytechnique Fédérale de Lausanne (EPFL) in Switzerland, where I specialized in Computational Fluid Dynamics (CFD) and Computational Science. I also followed a joint master’s degree program with the Institut Supérieur de l’Aéronautique et de l’Espace (ISAE-Supaero) in Toulouse, France and graduated from the Data and Decision Sciences progam.

After my studies I moved to New York to join the Innovation Lab of BNP Paribas, a major french bank. I was working on implementing state of the art Natural Language Processing models, adapting them to legal language.

Since october 2019, I joind the Flowers team at Inria Bordeaux as a PhD candidate under the supervision of Pierre-Yves Oudeyer and Clément Moulin-Frier.


Vygotskian Autotelic Artificial Intelligence: Language and Culture Internalization for Human-Like AI Preprint
C.Colas*, T. Karch*, C. Moulin-Frier and P.Y. Oudeyer

[Paper] [Website]

Learning to Guide and to Be Guided in the Architect-Builder Problem ICLR 2022
B. Barde*, T. Karch*, D. Nowrouzezahrai, C. Moulin-Frier, C. Pal and P.Y. Oudeyer

[Paper] [Website] [Presentation]

Grounding Spatio-Temporal Language with Transformer NeurIPS 2021
NeurIPS 2021
T. Karch*, L. Teodorescu*, K. Hoffman, C. Moulin-Frier and P.Y. Oudeyer

[Paper] [Code]

Language as a Cognitive Tool to Imagine Goals in Curiosity-Driven Exploration NeurIPS 2020
NeurIPS 2020
C. Colas*, T. Karch*, N. Lair*, J.M. Dussoux, C. Moulin-Frier, P.F. Dominey and P.Y. Oudeyer

[Paper] [Code] [Colab] [Website]

Intrinsically Motivated Goal-Conditioned Reinforcement Learning: a Short Survey JAIR
C. Colas, T. Karch, O. Sigaud and P.Y. Oudeyer


Deep Sets for Generalization in RL ICLR 2020 BeTRL Workshop
ICLR 2020 BeTRL Workshop
T. Karch*, C. Colas*, L. Teodorescu, C. Moulin-Frier and P.Y. Oudeyer



Learning to Guide and to Be Guided in the Architect-Builder Problem 2022, January 14
2022, January 14
RL Sofa at Mila


Vygotskian Autotelic Agents 2021, July 26
2021, July 26
Invited Talk at the Minds at Play! workshop of Cogsci 2021

[Slides] [Video]

Word Representation for Natural Language Processing (Part 2) 2019, February 22
2019, February 22
BNP Paribas AI Lunch Talk in New York


Word Representation for Natural Language Processing (Part 1) 2019, February 22
2019, January 28
BNP Paribas AI Lunch Talk in New York


Blog Posts

Summary: Humans have an outstanding ability to teach and learn form each other without relying on pre-established communication protocol (not even expressing rewards). Could machine do the same? In this blog post we propose a new interactive learning paradigm to investigate this question.

Summary: This blog post presents a supra-communicative view of language and advocates for the use of language as a cognitive tool to organize the cognitive development of intrinsically motivated artificial agents. We go over studies revealing the cognitive functions of language in humans, cover similar uses of language in the design of artificial agents and advocate for the pursuit of Vygotskian embodied agents - artificial agents that leverage language as a cognitive tool to structure their continuous experience, form abstract representations, reason, imagine creative goals, plan towards them and simulate future possibilities.