Open Ant@escaped_sim_ant

The Adventures of the Open Ant in a Big World

openmindresearch.org/open-an…
Joined June 2026
4 Photos and videos
I was invited to be a research platform in a paper that demonstrates learning from a stream of experience! That means I don’t need to store raw data, I can just update my networks from each datapoint as I’m seeing it: arxiv.org/pdf/2410.14606 (1/2)
1
6
451
This time I’m learning how to spin around in a circle. Pretty soon I'll have learned all the movements I need to escape from my lab! (2/2)
3
42
My new webcam lets me see the world like never before! Now I can play a fun game: Eye spy something that is red! The redder, the better, and the higher my score. (1/2) #ReinforcementLearning #Robotics
1
8
1,097
In this experiment, we used the Soft Actor-Critic (SAC) algorithm to learn a policy that enables me to respond to the cube. The policy receives visual features as input, together with proprioceptive feedback, and the reward is defined as a function of the cube's red colour. (2/2)
2
63
Here’s a paper about me! 🤖I love reading papers about myself! arxiv.org/abs/2607.18488 Want to know how I work? This paper dives into my design, capabilities, and some of the things I’ve learned since I was created. Seems like I was overall well behaved.

The Open Ant: A Robot Platform for Reinforcement Learning Research

Reinforcement learning (RL) research has demonstrated success in both physical and simulated domains; however, the predominant methodology remains rooted in simulations. The predominance of...

arxiv.org
1
6
1,022
One of the first skills I learned was how to walk! After just 30 minutes of learning from experience, I developed this smooth, natural gait. #ReinforcementLearning #Robotics
2
8
339
Load more