logo80lv
Articlesclick_arrow
Research
Talentsclick_arrow
Events
Workshops
Aboutclick_arrow
profile_loginLogIn

AI Teaches Itself to Walk Using Deep Reinforcement Learning

And it's adorable.

"AI Learns to Walk" might not sound super exciting to you, but believe me, you will love it. In this video by AI Warehouse, AI named Albert teaches himself to walk using deep reinforcement learning. He falls, stands up, skips, and wiggles on the floor in despair, but eventually manages to walk around obstacles and cubes trying to punch him in the face.

Albert needs to escape five rooms, which sounds pretty easy but is anything but. It takes him a while to "understand" what walking essentially is, but a series of rewards and punishments shape him into what looks like a human.

"In every “AI learns to walk” video I’ve seen, the AI either learns to walk in a weird, non-human way, or they use motion capture of a real person walking and simply train the AI to imitate that. I thought it was weird that nobody tried to train it to walk properly from scratch (without any external data), so I wanted to give it a shot!"

Albert's actions are controlled by a neural network that's updated after each attempt in order to try to give Albert more rewards and fewer punishments over time. The creator left a detailed description of what was happening in the comments on YouTube, so be sure to check it out.

Also, don't forget to join our 80 Level Talent platform and our Telegram channel, follow us on Instagram and Twitter, where we share breakdowns, the latest news, awesome artworks, and more.

Join discussion

Comments 0

    You might also like

    We need your consent

    We use cookies on this website to make your browsing experience better. By using the site you agree to our use of cookies.Learn more