François Chollet

François Chollet
Ramosset · CC BY-SA 4.0 · source
Name	François Chollet
Birth date	1985
Occupation	Researcher; Software Engineer; Author
Known for	Keras; Deep learning research; AI ethics
Employer	Google Research
Alma mater	École Polytechnique; Télécom Paris

François Chollet is a French software engineer, researcher, and author primarily known for creating the Keras deep learning library and for contributions to machine learning, computer vision, and artificial intelligence ethics. He has worked at major technology and research organizations and authored influential books and software that bridge academic research and practical engineering. Chollet's work sits at the intersection of applied machine learning, software engineering, and debates about general intelligence.

Early life and education

Chollet studied mathematics and computer science at institutions including École Polytechnique and Télécom Paris, and completed graduate work that connected to research communities associated with INRIA, CNRS, and French technological hubs near Paris. During his formative years he engaged with research groups linked to École Normale Supérieure and collaborations influenced by European initiatives such as Horizon 2020 and networks involving Imperial College London and ETH Zurich. His academic trajectory intersected with scholarship at institutions comparable to Massachusetts Institute of Technology, Stanford University, and University of Oxford through conferences and visiting collaborations.

Research and contributions

Chollet's research spans deep learning architectures, convolutional networks, representation learning, and evaluation of intelligence. He has published and implemented models related to convolutional neural networks used in work by groups at Google Brain, DeepMind, Facebook AI Research, and labs at MIT-IBM Watson AI Lab. His contributions include practical model-building patterns adopted across projects at Microsoft Research, Amazon Web Services, and academic centers such as Carnegie Mellon University and University of California, Berkeley. He has participated in conferences like NeurIPS, ICML, ICLR, and CVPR, and his code and writings influence benchmarks used by teams from OpenAI, Allen Institute for AI, Stanford Artificial Intelligence Laboratory, and Berkeley AI Research.

Chollet has explored evaluation frameworks for generalization and intelligence, dialoguing with scholars from DeepMind and critics at Harvard University, Princeton University, and Oxford University on notions of sample efficiency, transfer learning, and scale. His ideas intersect with historical work by researchers associated with Geoffrey Hinton, Yoshua Bengio, and Yann LeCun and contrast with perspectives from scientists at Searle-influenced philosophical traditions and cognitive science groups at MIT and University of California, San Diego.

Publications and software projects

He authored books that synthesize practice and theory for practitioners and researchers, publishing material that complements texts by authors affiliated with O'Reilly Media, MIT Press, and Cambridge University Press. His notable written work includes a book addressing artificial general intelligence debates and critiques of popular narratives, producing discourse alongside authors from Nick Bostrom-related circles at University of Oxford and commentators at Brookings Institution and RAND Corporation.

Chollet created the Keras library, a high-level neural networks API used by engineers and researchers at Google, Facebook, Microsoft, and startups in the Silicon Valley ecosystem, adopted in platforms such as TensorFlow and integrated with services from Google Cloud Platform and Amazon SageMaker. Keras has been used in projects spanning computer vision, natural language processing, and reinforcement learning alongside libraries like PyTorch, TensorFlow, MXNet, and JAX. He has released open-source code and model implementations adopted by contributors from institutions including University of Toronto, University of Montreal, and New York University.

Other projects and contributions include implementations of image classification models, utilities for model interpretability used in collaborations with researchers at UCL, Columbia University, and Johns Hopkins University, and tooling that interacts with datasets curated by ImageNet, COCO, MNIST, and research efforts at Allen Institute for AI.

Career and positions

Chollet has held roles at Google Research and collaborated with teams at Google Brain, engaging with engineers and scientists working on machine intelligence, cloud infrastructure, and applied research. He interacts with product and research teams similar to those at DeepMind, OpenAI, and industrial research labs at NVIDIA Research. His professional network includes contacts across academic departments at Stanford University, Harvard University, Princeton University, and industrial research groups at Apple, Amazon, and Facebook.

He has contributed to community education through workshops and tutorials at venues such as NeurIPS, ICML, CVPR, and summer schools organized by The Alan Turing Institute and European Laboratory for Learning and Intelligent Systems (ELLIS).

Awards and recognition

Chollet's work has been recognized in the machine learning community through citations, adoption of software, and invitations to speak at international conferences hosted by organizations like NeurIPS, ICLR, and ICML. His projects have been highlighted in technical coverage by outlets associated with IEEE, ACM, and professional societies including Society for Industrial and Applied Mathematics and recognition in developer communities around GitHub and Stack Overflow. He has been acknowledged in industry reports from groups such as Gartner and think tanks like MIT Technology Review.

Public engagement and influence

Chollet engages publicly through blog posts, code repositories, and talks that inform policy debates and technical directions, interacting with policymakers and commentators linked to European Commission, US National Institute of Standards and Technology, and advisory bodies around AI governance. His writings have influenced discourse shared with researchers from Oxford Internet Institute, Center for a New American Security, and commentators at Brookings Institution. Through open-source advocacy, he has shaped developer education used by students at universities including University of Cambridge, Imperial College London, and Tsinghua University.

Category:Machine learning researchers Category:French computer scientists