Searching...
14 results for “decision making”
Concept2 mentions across episodes
decision-making
Topic2 mentions across episodes
decision-making
task straight up, it might kinda struggle with it. So instead, you give it this task and you say, hey. Don't answer this. Before answering it, tell me what are some subproblems that would need to be solved first? And then it gives you a list of subpr
25:30 / 26:30
from transcript25:30 – 26:30
One way to test if it's giving you the right answer is if it's consistently doing the same thing that you could write a code to to tell you this is true or false. For example, will it ever say there's a virtual tour? So you could ask it, is do you pr
50:06 / 51:17
from transcript50:06 – 51:17
...can make a final decision about whether they can return it, if there's any charges, and that sort of thing. What is the phrase that you recommend people use? Is it what are the subproblems you need to solve first? Yeah. That that is the the phrasing
27:45 / 28:46
from transcript27:45 – 28:46
And that's It's a rare. It's a rare. Yeah. So, you know, as a product manager or someone even if you're not doing this calculation yourself, if someone ever reports to you agreement, you should immediately ask, okay. Tell me more. Like, you need you
59:15 / 1:00:19
from transcript59:15 – 1:00:19
especially in climate, especially in, kinda like, agriculture, food security, you can't think of this as, you know, like shots on goal and this and that. You've gotta kind of say, hey, we can get better at this. Reasoning is the biggest paradigm shif
10:27 / 11:44
from transcript10:27 – 11:44
And so that that's a resurgence of of that movement where we have this amazing generative, neural network that is out of the LLM. And now let's layer on more discrete ways of trying to verify whether it's doing the right thing or not, and let's put t
24:43 / 25:56
from transcript24:43 – 25:56
doesn't do, actually that well in GDP val, AA. Yeah. The the thing that you have to watch out for with Olympiadge is self preference that models usually prefer their own output. And in this case, it was not. Totally. I think the the the way that we'r
41:57 / 43:22
from transcript41:57 – 43:22
Vocabulary to plan over very long horizons and actually able to retain ability to retain memory. And back to eval's question, that's why I think eval's of the form of how long does this model autonomously operate for are of particular interest to us.
7:52 / 9:48
from transcript7:52 – 9:48
Now the human trainer already knows what the solution would look like. So we have a pull request that we have on GitHub, so we know exactly, or we have a unit test that we can run and verify the solution. So what it does is it rolls out a lot of diff
15:43 / 16:49
from transcript15:43 – 16:49
And, in fact, I think, like, with the reasoning models, we have seen the models, like, greatly extend the length over which they are able to reason and work reliably without going off track. Yeah. I think this has remained a big Yeah. Area of focus f
8:59 / 10:00
from transcript8:59 – 10:00
And as the code becomes bigger or more complex or more input, it's gonna just start kind of sloshing back and forth between bugs. And so if you don't have those hard skills and you're not ultimately the driver at the end of the day, like, you're gonn
4:38:58 / 4:40:00
from transcript4:38:58 – 4:40:00
...of reasoning is making models better with thinking and post training. K? Yeah. So basically, like, RL ing the model to think better. Right? And thinking is more like thinking traces and thought trajectories and stuff like that. Right? There's also th
32:53 / 34:00
from transcript32:53 – 34:00
Have a podcast?
Run your podcast through Clypt
Get ranked clips, hooks, and ready-to-post copy from your own episodes. Free to try.