text stringlengths 0 2.68k |
|---|
**Ewelina Kurtys:** Absolutely. |
**Jerod Santo:** And so, of course, it's going over copper wires, and Wi-Fi, and whatever backbones, and then back into your interface, which eventually translates it into... I'm imagining there's a needle at the end of a thing that like sprays some dopamine... I don't know what happens. What happens at the end, the la... |
**Ewelina Kurtys:** Yes. So when you send an electrical signal, you have a digital to analog converter. So you have to translate the things from digital world to analog world, because neurons are analog. |
**Jerod Santo:** Right. |
**Ewelina Kurtys:** So you have this digital to analog converter, and it is translated into electrical signals, which goes through the electrodes. So basically, the stream of electrons are going, flowing through the electrodes. And when you want to send a signal with dopamine or serotonin, then it's connected to the la... |
**Jerod Santo:** Okay. |
**Ewelina Kurtys:** And when the UV light is open, the dopamine is released... Because dopamine is closed chemically. It's encaged in the chemical, so that it's not active. But when it sees the UV light, then it's released. |
**Jerod Santo:** Oh... |
**Ewelina Kurtys:** So it's a way of releasing very, very quickly dopamine to the neurons. So this is how it works. So then it's connected to some controllers, which are connected to the lamp, and then the lamp switch on. |
**Jerod Santo:** Okay. Same thing for serotonin? |
**Ewelina Kurtys:** Yes. But for serotonin, we have different wavelengths. I don't remember which one. It's not UV. But then we have different wavelengths, yes... So that they don't overlap. So you can have both in the medium. And of course, we plan to have more of this, and of course, much more neurotransmitters... Bu... |
**Adam Stacoviak:** Are those hormones? Are those chemicals? What is the proper terminology to call dopamine and serotonin? |
**Ewelina Kurtys:** No, they're neurotransmitters. So you have several in the brain, and they affect learning. And actually, the whole idea of using them is because we use them for feedback. Because, actually, the way how humans are learning is by feedback. So you have interaction with the environment, you get feedback... |
\[32:02\] So the idea - at least our idea here, because it's a bit complicated... And actually, there are different opinions about how to give punishment or reward to neurons... But our idea is to give dopamine as a reward. And no dopamine as a punishment. So that's used for the feedback loop. So for example, you stimu... |
**Adam Stacoviak:** And these neurons learn. |
**Ewelina Kurtys:** Well, that's the problem - sometimes they learn, sometimes they don't. This is still a challenge, you know. This is still a challenge, learning. So learning for neurons is changing the connections, changing the behavior of neurons, so behavior will be electrical activity... And this is still a chall... |
**Break**: \[33:04\] |
**Adam Stacoviak:** Remind me... Dopamine is positive, so that's used for rewards... What is serotonin used for? What do the two levers do? |
**Ewelina Kurtys:** No, actually, that would also be for the -- |
**Jerod Santo:** Two different versions of reward. |
**Ewelina Kurtys:** Yes. However, if we have to go to the details, it's a little bit tricky... So we are still -- |
**Adam Stacoviak:** Okay... Let's go into the details. Let's get tricky. |
**Ewelina Kurtys:** \[36:11\] Yes. Because actually, dopamine is -- it also depends when it is given, and also there are different receptors. So receptors are on the surface of the cells. So if they have dopamine receptors, that means they can recognize dopamine, actually, because they need always a receptor to recogni... |
**Adam Stacoviak:** So it's not stable yet. |
**Ewelina Kurtys:** No. |
**Adam Stacoviak:** Some days they do, and some days they don't... And you're learning. So this is 10 years before it's usable in production. This is total lab learning... What is it that -- and maybe you're still early and you can't answer this question, but what is it that makes it such a variable? Is it just because... |
**Ewelina Kurtys:** Yes, because it's bio. Because first -- there are two reasons, I would say, main. First is because it's bio, so it's unstable. Second is because nobody knows yet how neurons encode information. This is totally different than digital. So this is such a challenge, because you have to understand a new ... |
**Jerod Santo:** But you do have some indicators that they at least generally work the same. |
**Ewelina Kurtys:** Well, you mean neurons? |
**Jerod Santo:** Yeah. Well, if each neuron was like a snowflake - you know, every snowflake is unique, and it melts, so it changes... Then there really would be no -- like, 10 years, 100 years, a billion years, there would be no getting there, because there's no determinism at all. Because everyone could just work com... |
**Ewelina Kurtys:** You know, I think it is deterministic. It's just that we don't know the rules yet. |
**Jerod Santo:** Right. That's why I'm saying, you do have an indicator that they do kind of work the same generally, though... At least for this one thing. |
**Ewelina Kurtys:** Yes. But the indicator is our brains, actually... Because we have no doubts that neurons can process information very well. This is why we can talk. |
**Jerod Santo:** True. |
**Ewelina Kurtys:** So we can say nature is a proof that neurons are working. |
**Jerod Santo:** That's fair. |
**Ewelina Kurtys:** There is no doubt about this. |
**Jerod Santo:** We think differently, but we all think, I guess... |
**Ewelina Kurtys:** We just have to learn how to program them. |
**Jerod Santo:** Have you tried telling them to ultrathink? I'm sorry, that was a joke... |
**Adam Stacoviak:** \[laughs\] Good throwback. |
**Ewelina Kurtys:** That was from a previous show. |
**Adam Stacoviak:** I have been using that, by the way... I've now said -- sorry for a slight aside... "Triple check and ultrathink." That's my new phrase. |
**Jerod Santo:** That's the key word, Ewelina, for certain AIs... |
**Adam Stacoviak:** Triple check that stuff and ultrathink. |
**Jerod Santo:** When they're not thinking best... But of course, with a neuron, maybe you just keep it analog and just whisper to it. "Ultrathink...!" Just walk up to it and whisper. |
**Ewelina Kurtys:** Yes, you can whisper, but they don't have ears, so they can only understand -- |
**Jerod Santo:** Well, you need some ear cells. Get some ear cells going. |
**Ewelina Kurtys:** You have to translate, because usually our ears also translate to electrical signals. |
**Jerod Santo:** True. |
**Ewelina Kurtys:** That's why our neurons in the head can understand. So you have to learn. That's the whole point - how to encode information so that they can understand. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.