Johan S. Obando-Ceron

I am a research assistant and master's student at Universidad Autonoma de Occidente (UAO) under the supervision of Victor Romero Cano with a focus on machine perception, deep learning, reinforcement learning and robotics. During my master's studies I have been working on sensor data fusion (Camera and LiDAR) for improved object recognition in the context of environment perception from mobile platforms.

Prior to that, I had a great fortune working with Ricardo Moreno Chuquen on problems of optimization and flexibility of modern power systems. I completed my undergraduate degree at the UAO where I studied Mechatronics Engineering. During my undergraduate study, I worked with Professor Jesus Alfonso Lopez as an undergraduate researcher in the Artificial Intelligence team.

My research interests span computer vision, reinforcement learning, and robotics. I am highly motivated to work in how algorithms can enable machines to acquire more general notions of intelligence through learning and interaction, allowing them to autonomously learn a variety of skills in real-world settings and build on top of already existing skills to solve more complicated tasks. While the successes of reinforcement learning in the single-task setting are incredibly exciting, I am most drawn to problems that stand in the way of efficient multi-task and lifelong reinforcement learning.

Frequently, I am studying on several topics related to RL, deep learning, cognitive-AI, multimodal learning and causality. Therefore, I would like to share to you some amazing academic sources available on internet. They have been very useful to me.

Thanks for visiting!

Email: jobando0730[AT]

GitHub / Google Scholar / Blog / LinkedIn / Twitter

  • [Aug 2020] I have been accepted into the Neuromatch Academy (online school for Computational Neuroscience) as an interactive track, 2020.
  • [Jun 2020] I will be starting my second master studies at Mila/UdeM in Computer Science, 2020.
  • [Oct 2019] I have been awarded a travel grant to attend the Conference on Neural Information Processing Systems NeurIPS, 2019. Vancouver, Canada.
  • [Oct 2019] The abstract "Exploiting the potential of deep reinforcement learning for classification tasks in high-dimensional and unstructured data" was accepted to the LatinX in AI Research Workshop at NeurIPS, 2019. Vancouver, Canada.
  • [Oct 2019] The abstract "Divide and Conquer: an Accurate Machine Learning Algorithm to Process Split Videos on a Parallel Processing Infrastructure" was accepted to the LatinX in AI Research Workshop at NeurIPS, 2019. Vancouver, Canada.
  • [Aug 2019] I have been awarded with transport, registration fee and accommodation scholarships to attend to the Latin American Meeting In Artificial Intelligence, Khipu.
  • [Aug 2019] I have been accepted at the Latin American Meeting In Artificial Intelligence - Khipu. Montevideo, Uruguay.
  • [Jun 2019] Presented my work of Probabilistic Perception System for Object Classification Based on Camera - LiDAR Sensor Fusion at the Latinx in AI Workshop @ ICML 2019 as poster, Long Beach, CA.
  • [May 2019] International Conference on Machine Learning (ICML) Travel Award.
  • [Jan 2018] Academic Excellence Scholarship: 100% scholarship to cover the full tuition cost for the whole MSc graduate program.
Blog Posts
  • The Option-Critic Architecture Pierre-Luc Bacon, Jean Harb, Doina Precup [arxiv][block]
  • Safe Option-Critic: Learning Safety in the Option-Critic Architecture [arxiv][block]
Source stolen from here