Home▸AI/ML Drug Discovery Infrastructure▸REINVENT: Policy-Gradient Molecule Generation

REINVENT: Policy-Gradient Molecule Generation

Watch a REINVENT-style policy-gradient agent build drug-like molecules token by token in 3D, scored against a target pocket size and heteroatom balance, and reinforced with the REINFORCE gradient over training batches.

AI/ML Drug Discovery Infrastructure3DAdvanced60 FPS

A compact policy learns to grow drug-like molecules one atom at a time — carbon, nitrogen or oxygen, branching, ring closure or stop — the same token-by-token generation REINVENT uses on SMILES strings. Every finished molecule is scored against a target pocket size and heteroatom balance, and the REINFORCE policy gradient pushes the token probabilities toward the choices that produced above-average reward. Watch the 3D molecular graph grow live, batch by batch, as the reward history climbs and the policy's action distribution sharpens.

⚙ Under the hood

Watch a REINVENT-style policy learn to grow drug-like molecules token by token in 3D, scored against a target pocket size and heteroatom balance and reinforced batch by batch with the REINFORCE policy gradient.

reinforcement learningdrug discoverymolecular generationREINVENTpolicy gradientThree.js

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install