arg min
Subscribe
Sign in
Home
Lecture Blogs
Collections
Archive
About
Latest
Top
Discussions
Benchmark Studies
It is impossible to disentangle technical innovation from technical debt
10 hrs ago
•
Ben Recht
16
11
3
There is no data-generating distribution
Reflecting on teaching machine learning again. Again.
Dec 9
•
Ben Recht
29
19
4
There's got to be a better way!
From Reformist RL to the principle of certainty equivalence.
Dec 5
•
Ben Recht
33
22
2
Defining Reinforcement Learning Down
It's a lot simpler than I realized.
Dec 3
•
Ben Recht
41
16
3
Random Search for Random Search
Digging into specific applications of policy gradient
Dec 2
•
Ben Recht
18
3
Reformist Reinforcement Learning
What if we just begin and end with policy gradient?
Dec 1
•
Ben Recht
44
21
4
November 2025
Digitally Twinning
Justifying generative modeling and the principle of maximum likelihood
Nov 20
•
Ben Recht
32
12
1
Actions from predictions
A general framework for outcome maximization. Is this reinforcement learning?
Nov 18
•
Ben Recht
13
4
How to pick a sample size.
A brief introduction to adaptive experimentation without saying "exploration-exploitation tradeoff."
Nov 13
•
Ben Recht
28
8
The DOI Directorate
Articulating a few concrete positions on archives, surveys, and position papers
Nov 12
•
Ben Recht
22
6
2
A position on positions
The complex evolution of academic process doesn't always lead us to better practice.
Nov 10
•
Ben Recht
28
23
1
Staging Interventions
Actions are fundamentally different than predictions, but it's hard to write this distinction in math.
Nov 6
•
Ben Recht
19
21
1
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
