Phil

Reinforcement Learning. Large Language Models. Freelancing.

[Phil @MLWithPhil]

@adammcdan @1337hero I'm not sure about the benefit of ECC for your use case, but you can always power limit the 5090 to 400-450W to mitigate power consumption, for not much performance hit.

[Phil @MLWithPhil]

@MasterTimBlais Who said Rick Rolling couldn't be done in 2026?

[Phil @MLWithPhil]

@sam_gzstrength This is timely. I just tweaked my back yesterday. Gotta get some barbell medicine.

[Phil @MLWithPhil]

@eurofounder This really is top tier bait.

[Phil @MLWithPhil]

@vetbosss @LundukeJournal Got a link to this? There seems to be some debate in other comments.

[Phil @MLWithPhil]

@John3Sobieskii @StefanMolyneux They debated many years ago, and Jay has more recently hosted Stefan on Infowars.

[Phil @MLWithPhil]

@MrVladimirX So many people missed the fundamental point. Society is about to bifurcate; those with access to super intelligence and those without.

The technical details are irrelevant and, at best, push the timeline by a few years.

[Phil @MLWithPhil]

@DigitalDavidREV @sam_gzstrength You're going to want to put on some weight. I don't think 200 is good at 5'9, but 155 is too light.

[Phil @MLWithPhil]

@greek_groyper @StefanMolyneux This hasn't been my personal experience. Our church has seen several young men join recently. If anything there's a shortage of women.

[Phil @MLWithPhil]

@yacineMTB 6 months of study under a priest; attending liturgy. Baptism. Fasting half the year, confession, and multiple services a week. Prayer is twice a day. Alms giving is as you can do it. Tithe if possible. Eastern Orthodoxy is the way.

[Phil @MLWithPhil]

@wojakcodes MERN?

[Richard Sutton @RichardSSutton]

My acceptance speech at the Turing award ceremony:

Good evening ladies and gentlemen.
The main idea of reinforcement learning is that a machine might discover what to do on its own, without being told, from its own experience, by trial and error. As far as I know, the first person to propose this was Alan Turing in 1947, which makes it particularly gratifying and humbling to receive this award in his name for reviving this essential but still nascent idea.
I have three people that I would like to particularly thank.
First, Andy Barto. As my PhD supervisor he taught me my whole approach to science, and in particular instilled in me an appreciation of scholarship and craft, and of the great breath of prior work.
Second, I would like to thank Oliver Selfridge, my other main mentor; sadly, now deceased. Oliver taught me how keeping ideas simple can be the boldest of all ambitions.
Third, I want to thank Martha Steenstrup, my life partner and intellectual sparring partner. She keeps me honest and grounded.
Finally, I also want to thank the University of Alberta, which has been an ideal environment for me and for reinforcement learning research these past 22 years.
These three people and my university have reinforced in me the ambition to have ideas that matter, without getting too full of myself about it. They taught me that the quest for better ideas is serious, but is best approached playfully, with humility, kindness, and optimism. For this I am eternally grateful.
I would also like to thank all of you for being here and for celebrating the pursuit of intellectual excellence.
Thank you very much.

[Phil @MLWithPhil]

Excited to dig into "Tiny C Projects" by @dangookin and @ManningBooks. Low level programming is the future, contrary to what some may have you believe.

[Richard Sutton @RichardSSutton]

I was happy to give a more technical talk on how we might create an AI at RLC-2025 and AGI-2025 (video below).
The Oak Architecture: A Vision of Super-Intelligence from Experience
As AI has become a huge industry, to an extent it has lost its way. What is needed to get us back on track to true intelligence? We need agents that learn continually. We need world models and planning. We need knowledge that is high-level and learnable. We need to meta-learn how to generalize. The Oak architecture is one answer to all these needs. It is a model-based RL architecture with three special features: 1) all of its components learn continually, 2) each learned weight has a dedicated step-size parameter that is meta-learned using online cross-validation, and 3) abstractions in state and time are continually created in a five-step progression: Feature Construction, posing a SubTask based on the feature, learning an Option to solve the subtask, learning a Model of the option, and Planning using the option’s model (the FC-STOMP progression). The Oak architecture is rather meaty; in this talk we give an outline and point to the many works, prior and contemporaneous, that are contributing to its overall vision of how super-intelligence can arise from an agent’s experience.
https://t.co/kTTGUavbIe

[Phil @MLWithPhil]

@StefanMolyneux Coherence

[Phil @MLWithPhil]

@BaronDestructo Dark Matter, of course. I'm still bitter over that cancellation.

[Santiago @svpino]

I recently came across a team using an LLM to implement a state machine.

They called it "agentic". They also told me it worked correctly most of the time.

I asked them to write down the list of rules for moving from one state to another.

They listed them out.

It didn't take long for them to realize what was happening: They had built a solution using AI for a problem that didn't need AI.

We removed the model and wrote some code to implement the rules. It might have taken 2-3 hours at most.

They went from "it works most of the time, it's relatively fast, and it costs some tokens" to "it works 100% of the time, instantaneous, and costs nothing."

So many examples like this recently.

Golden rule: BUILD THE SIMPLEST THING THAT COULD POSSIBLY WORK.

[John Carmack @ID_AA_Carmack]

It was a fun exercise recently to just open up a completely blank file and write an RL agent from scratch, without looking at any of my prior code. There is a point of scale where rewriting things from scratch is a bad idea, but it is a blessing when you can!

By “from scratch” I mean with pytorch, and I chuckle a bit about how I used to be irritated when people would say that, as I considered it not from scratch if it wasn’t in C with no libraries, but now I am them.

[Phil @MLWithPhil]

@Srini_Pa @GaryMarcus That's great and all, but it does nothing to address the conversation at hand.