Interaction-Grounded Learning: Learning from feedback, not rewards

The IGL Setting

Taken from Xie et al., 2021
Code available here.
Learned and true reward achieved over time. X-axis corresponds to epochs. Y-axis corresponds to achieved reward.

Applying IGL to “Real Problems”

--

--

--

Research Scientist. Interested in Artificial Intelligence, Neuroscience, Philosophy, and Literature.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Effective TensorFlow for Beginners

Machine Learning on Graphs @ NeurIPS 2019

R Pokemon Legendary?

A hands-on intro to ANNs and deep learning — part 1

Automated Text Analysis using Streamlit

Fake News Detection

FixMatch: A Semi-Supervised Learning method,that can be learned even if there is only one labeled…

Machine learning on a budget: AI for small businesses

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Arthur Juliani

Arthur Juliani

Research Scientist. Interested in Artificial Intelligence, Neuroscience, Philosophy, and Literature.

More from Medium

Counterfactuals for Reinforcement Learning II: Improving Reward Learning

Oxford U Proposes COIN++, a Neural Compression Framework for Different Data Modalities

#6: RL Goes Nuclear, RL Assists Anesthesiologists, MuZero Powers YouTube, Outracing Champion Gran…

RL — Guided Policy Search (GPS)