Wednesday, November 4, 202625 min read
Courses · Reinforcement Learning
ε-greedy control settles for the best policy that keeps exploring
Monte Carlo control with ordinary starts needs a policy that sometimes moves at random, and the price of those moves grows with ε, first in value and then in the learned arrows.
Daily · Feature
Muon's fine-tuning lead runs out after a similar token count at every size tested
Across a CMU preprint's 30M-to-600M models, AdamW pretraining takes the lead between 2.4 and 19.2 billion tokens, and production models pretrained with Muon have seen trillions.
Daily · Daily Brief
Democrats win back the House and the Senate
By AP’s 1:35 a.m. Eastern count, Democrats had gained 25 House seats and seven Senate seats, and Missouri voters had rejected the map Trump asked for.