People
  1. Academic experts
  2. Research students
  3. Students
  4. Alumni
  5. Senior people at City
  6. Non-academic staff
  7. Honorary graduates
People

portrait of Michael Garcia Ortiz

Michael Garcia Ortiz

Lecturer in Artificial Intelligence

School of Mathematics, Computer Science and Engineering, Department of Computer Science

Contact Information

Contact

Visit Michael Garcia Ortiz

A309F, College Building

Postal Address

City, University of London
Northampton Square
London
EC1V 0HB
United Kingdom

About

Overview

My main interest lies in Artificial General Intelligence and Robotics, and in particular the emergence of common-sense knowledge through sensorimotor prediction. Sensorimotor prediction is a very powerful drive, allowing an embodied agent to acquire knowledge and build a model about its world. I am particularly interested in end-to-end self-supervised learning approaches inspired from insights in Neuroscience and Cognitive Development. Recently, I have been working on topics involving agents navigating in a physical world and learning representations of spatial displacement, as well as learning the structure of their environment (emergence of object representation), guided solely by prediction on continuous sensorimotor streams.

List of topics of Interest:
- Artificial General Intelligence
- State Representation Learning
- Sensorimotor Prediction
- Continual Learning

I am a member of the Research Center on Artificial Intelligence CitAI:
CitAI

Employment

  1. Lecturer, City, University London, Jul 2019 – present
  2. Research Scientist, Softbank Robotics Europe, Sep 2013 – May 2019

Publications

Conference papers and proceedings (10)

  1. Caselles-Dupré, H., Garcia Ortiz, M. and Filliat, D. (2019). Symmetry-Based Disentangled Representation Learning requires Interaction
    with Environments.
    Thirty-third Conference on Neural Information Processing Systems (NeurIPS 2019) 8-14 December, Vancouver, Canada.
  2. Laflaquiere, A. and Garcia Ortiz, M. (2019). Unsupervised Emergence of Egocentric Spatial Structure from Sensorimotor Prediction. Thirty-third Conference on Neural Information Processing Systems 8-14 December, Vancouver.
  3. Caselles-Dupré, H., Garcia Ortiz, M. and Filliat, D. (2019). S-TRIGGER: Continual State Representation Learning via Self-Triggered
    Generative Replay.
  4. Lesort, T., Caselles-Dupré, H., Garcia Ortiz, M., Stoian, A. and Filliat, D. (2019). Generative Models from the perspective of Continual Learning. International Joint Conference on Neural Networks (IJCNN) 2019.
  5. Caselles-Dupré, H., Garcia Ortiz, M. and Filliat, D. (2018). Continual State Representation Learning for Reinforcement Learning using
    Generative Replay.
    Workshop on Continual Learning, NeurIPS 2018- Thirty-second Conference on Neural Information Processing Systems.
  6. Garcia Ortiz, M. and Laflaquiere, A. (2018). Learning Representations of Spatial Displacement through Sensorimotor Prediction. International Conference on Development and Learning and on Epigenetic Robotics (ICDL-EpiRob).
  7. Caselles-Dupre, H., Annabi, L., Hagen, O., Garcia Ortiz, M. and Filliat, D. (2018). Flatland: a Lightweight First-Person 2-D Environment for Reinforcement Learning. Workshop on Continual Unsupervised Sensorimotor Learning - ICDL-Epirob 2018.
  8. Kulak, T. and Garcia Ortiz, M. (2018). Emergence of Sensory Representations Using Prediction in Partially Observable Environments. Artificial Neural Networks and Machine Learning - ICANN 2018.
  9. Annabi, L. and Garcia Ortiz, M. (2018). State representation learning with recurrent capsule networks. Workshop on Modeling the Physical World: Perception, Learning, and Control, NeurIPS 2018 - Thirty-second Conference on Neural Information Processing Systems.
  10. Garcia Ortiz, M. (2017). Sensorimotor Prediction with Neural Networks on Continuous Spaces. Artificial Neural Networks and Machine Learning – ICANN 2017.

Journal articles (5)

  1. Gepperth, A.R.T., Garcia Ortiz, M., Sattarov, E. and Heisele, B. (2016). Dynamic attention priors: a new and efficient concept for improving object detection. Neurocomputing, 197, pp. 14–28. doi:10.1016/j.neucom.2016.01.036.
  2. Laflaquière, A. and Ortiz, M.G. Unsupervised Emergence of Spatial Structure from Sensorimotor Prediction. .
  3. Laflaquière, A., Hemion, N., Ortiz, M.G. and Baillie, J.-.C. Grounding Perception: A Developmental Approach to Sensorimotor
    Contingencies.
    .
  4. Laflaquière, A. and Ortiz, M.G. Unsupervised Emergence of Egocentric Spatial Structure from Sensorimotor
    Prediction.
    .
  5. Kulak, T. and Ortiz, M.G. Representation Learning in Partially Observable Environments using
    Sensorimotor Prediction.
    .