wegfawefgawefg

34 ms·

421.
by wegfawefgawefg 3y ago
i think he just means the example code defines components, but doesnt have a entity modification code with queries and stuff. im pretty used to existing ecs libs and they all let you add entities as collections of components and do queries.
422.
by wegfawefgawefg 3y ago
to tone down your enthusiasm a bit, heuristic tree search and q learning were already mixed years ago. Thats what Alphago and Alphazero were.
423.
by wegfawefgawefg 3y ago
the ml method doesnt require you to know how to solve the problem at all, and could someday presumably develop novel solutions. not just high efficiency symbolic graph search.
424.
by wegfawefgawefg 3y ago
As an ML programmer, i think that approach sounds really too complicated. It is always a bad idea to render the output of one neural network into output space before feeding it into another, rather than have them communicate in feature spac
425.
by wegfawefgawefg 3y ago
The real world is a space of continuous actions. To this day Q algorithms have been ones of discrete action outputs. I'd be surprised if a Q algorithm could handle the huge action space of language. Honestly its weird they'd consi
426.
by wegfawefgawefg 3y ago
Your comment is regarding LLMs, but Q* may not refer to an LLM. As such, our intuition about the failure of LLM's may not apply. The name Q* likely refers to a deep reinforcement learning based model. To comment, in my personal experie
427.
by wegfawefgawefg 3y ago
I do not think that is correct as the RL in RLHF already stands for reinforcement learning. :^) However, I do think you are right that self play, and something like reinforcement learning will be involved more in the future of ML. Tradition
428.
by wegfawefgawefg 3y ago
I could have given this lecture, and I think I could have made it much more entertaining, with fun examples. Lex should stick to what he likes, though his interviews can be somewhat dull. On occasion I learn things from his guests I would h
429.
by wegfawefgawefg 3y ago
Given the RLHF post training, I do believe it was intentional. And I suspect there have been iterations on this to make it more "robust". I vaguely remember there being announcements and such.
430.
by wegfawefgawefg 3y ago
When I write raytracing code or fiddly logic for games it will often give me comments along the lines of #I have no idea how this will work, probably should look this up.
431.
by wegfawefgawefg 3y ago
To give context on this video for anyone who doesn't understand. In this video PI* is referring to an idealized policy of actions that result in the maximum possible reward. (In Reinforcement Learning PI is just the actions you take in
432.
by wegfawefgawefg 3y ago
If I had to guess, the name Q* is pronounced Q Star, and probably the Q refers to Q values or estimated rewards from reinforcement learning, and the star refers to a search and prune algorithm, like A* (A star). Possibly they combined deep
433.
by wegfawefgawefg 3y ago
I dont think the theories here about why chatgpt puts out such bland content are correct. I don't think it is bland due to an averaging effect of all the data. The reason I dont think that is the case: I used to play with GPT3 and 3 wa
434.
by wegfawefgawefg 3y ago
Yes of course I have. I always tell them im looking for work lol. They just arent in a position to hire me. I had a meeting four days ago with one, who has a small startup, but very low funds. He said he will check back with me in about a m
435.
by wegfawefgawefg 3y ago
I have had handfuls of university students come to me for help in the full pocess of deciding what to make a paper about, coding the architectures for the paper, and reviewing the paper. That includes masters papers,and it has sorely affect
436.
by wegfawefgawefg 3y ago
I think I was getting filtered out before regardless. Probably by machine.
437.
by wegfawefgawefg 3y ago
A lot of it was contract work. Some of it I didnt even have a contract. I have no agency. If i removed my gigs less than a year then my resume is going to be almost blank. Secondly, on the ML front. I didnt just pick it up. Firstly, I pract
438.
by wegfawefgawefg 3y ago
That is reasonable advice. Thank you. I had it reduced to all back end development for a few months. I got no responses. To be honest I've given up on it.
439.
by wegfawefgawefg 3y ago
It has been circumstance. When you work for small companies there are sometimes problems. At the BP subcontractor there was the OPEC collusion that year. I saw an entire floor of Westlake campus get fired. I was a contractor, and I started
440.
by wegfawefgawefg 3y ago
-I dropped out of university because i wasn't allowed to take only compsci and math. 80% of my first two semesters were pointless state required classes so i left. That was 11 years ago. I love programming and I do it first thing in th
441.
by wegfawefgawefg 3y ago
I launched a rocket in factorio. In seriousness, I have hardware test and yield assessment in my background. Ive been to multiple pcb lines, talked to techs, worked with electrical engineers on pointing out repeated failures in tombstoned r
442.
by wegfawefgawefg 3y ago
As for my friend developers. One works at apple, one works at amazon, those are unlikely hires for me. One got laid off this year and now is getting certified to drive a city bus. A few of them work for the BP subcontractor, which is curren
443.
by wegfawefgawefg 3y ago
I have only worked for small companies. When you work for small companies you have to wear hats. When I worked at skylark I had to write cli tools to interact with their SoC in c and bash, but also do hardware test in python, but also make
444.
by wegfawefgawefg 3y ago
No im genuinely having a difficult time getting any responses at all. I've had three interviews since march. Method of application doesnt seem to matter, LinkedIn or Company Site, etc. I am not entry level. I have seven years of profes
445.
by wegfawefgawefg 3y ago
comma does raw input to output behaviour cloning like you propose. interestingly, tesla does not. via there last public ml update, theyre self driving architecture still involves a path generator and evaluator combo, and a lot of seperatel
446.
by wegfawefgawefg 3y ago
https://learndrl.com I wrote this extensive tutorial for teaching deep reinforcement learning, with a focus on getting intuition from code. you will find RL theory is heavy on math despite needing math for very little other than
447.
by wegfawefgawefg 4y ago
wealth and social status could be considered proxies for iq (g)
448.
by wegfawefgawefg 4y ago
web3 dev here. nft contracts are turing complete. you can put whatever code you want in them. usually people inherit the openzeppelen base nft contract which doesnt have royalties turned on by default (perhaps new version has the functional