Skip to content
View Josh012006's full-sized avatar
🔎
Exploring
🔎
Exploring
125 followers · 63 following

Achievements

Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: YOLO

Achievements

Achievement: Pull Sharkx3Achievement: QuickdrawAchievement: YOLO

Highlights

  • Pro

Block or report Josh012006

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Josh012006/README.md

Hi there 👋🏼, I'm Josué

Passionate about autonomous learning, robotics and molecular computing

I am an intelligent systems developer with a passion for reinforcement learning, robotics, and their applications to the biomedical field. I am interested in the different facets of reinforcement learning and its use in training autonomous agents, as well as in exploring both medical robotics and the field of molecular robotics. In particular, I am interested in how reinforcement learning can advance robotic applications in medicine, from large-scale robotic systems to micro- and nanorobotics.

Most of my recent work revolves around designing custom reinforcement learning pipelines, investigating the theoretical properties of RL algorithms, studying autonomous agents' learning dynamics, working with vision-language-action models (VLAs), and exploring the possibility of encoding reinforcement learning in living matter.

My research interests include sequential and long-horizon decision-making, deep reinforcement learning, goal-conditioned RL, inverse reinforcement learning (IRL), and algorithm optimization. I am particularly interested in understanding how these approaches can improve the learning and decision-making capabilities of autonomous agents. From a more biologically oriented perspective, I’m also drawn to DNA neural networks and chemical reaction networks, particularly their potential as alternative computational paradigms.

An RL agent going from random exploration to the shortest path

🚀 I'm currently working on autonomous control and navigation with goal-condition reinforcement learning in my project [NanoGoal-RL]. I've been sharpening my skills in algorithm optimization through [PPO-Leftovers]. And in parallel, I'm exploring how I can recover some kind of reward Caenorhabditis elegans is optimizing in [Swim-IRL].

I have many more projects that I pinned just below. Check them out and let me know what you think. Also feel free to contribute or leave a star if you appreciate my work !

You can contact me via :

💪 Skills:

I have experience working with reinforcement learning frameworks like Gymnasium and Stable-Baselines3. I understand and I can write code in Python and C/C++. On the engineering side, I am well-versed in Linux/Bash. Most of my work is assisted by coding agents. So I also know how to use them efficiently and in a secure way to learn and to work.

Apart from that, I have 2+ years of experience in front end web development and a bit of experience when it comes to the back-end. So I can also design React and Vue web applications.

🌟 What I'm Looking For:

I'm up to work on exciting projects that align with my interests.

🎉 Fun Facts about me:

When I'm not coding, I like to play a bit of 🎹 piano (I'm not that good though 😅) or watching japanese animation pieces (animes) 🇯🇵 🈺. My favorite anime is called Code Geass, give it a try! Also I'm a christian always trying to bring 👑 Jesus's peace and love everywhere I go. I'm particularly fond of doing things in a simple way. And lastly, I enjoy teaching and sharing my knowledge with others.

My activity reports

commit stats activity graph


commit stats commit stats

Pinned Loading

  1. CIB-IRL CIB-IRL Public

    Unsupervised image clustering via simultaneous Information-Bottleneck representation learning and Multi-Intention Inverse Reinforcement Learning. Listed on XERJ Corpus Hub.

    Python 1

  2. NanoGoal-RL NanoGoal-RL Public

    A goal-conditioned reinforcement learning project where a simulated 2D nanorobot learns to reach various target positions, showcasing autonomous decision-making, trajectory optimization, and RL in …

    Python 5 2

  3. OpenX-Embodiment-Datasets-Visualization OpenX-Embodiment-Datasets-Visualization Public

    Visualize end-effector spatial distributions and task-description semantic embeddings across 50+ Open X-Embodiment datasets. Built to identify OOD task candidates for RL fine-tuning of OpenVLA and …

    Jupyter Notebook 1

  4. PPO-Leftovers PPO-Leftovers Public

    What is Proximal Policy Optimization (PPO) leaving behind : a study of PPO's exploitation mechanism.

    Python 1

  5. Swim-IRL Swim-IRL Public

    Inverse reinforcement learning on real trajectories of active microscopic swimmers (starting with C. elegans chemotaxis), to infer the reward function they behave as if optimizing — MaxEnt IRL, Dee…

    Python 3

  6. IFT2255-ma-ville-bugbusters IFT2255-ma-ville-bugbusters Public

    An app MaVille to help the city of Montreal manage the public works more efficiently. It uses Java for the backend, React for the frontend and Morphia (MongoDB) for the database management. Built a…

    HTML 2 1