Dynex：神经形态量子计算云解决人工智能、量子计算和研究中的现实问题

人们仍在尝试构建更好的区块链并定制他们的网络以解决更多最新的计算挑战。其中之一是 Dynex

Dynex is a blockchain that was launched in 2022. It is aiming to solve “real-world problems” in AI, quantum computing, and research. It is described by its developers as a “distinctive, adaptable blockchain system” and “the world’s only accessible neuromorphic quantum computing cloud.” It harnesses the parallel processing capacity of its GPU-based mining network.

Dynex 是于 2022 年推出的区块链。它的目标是解决人工智能、量子计算和研究中的“现实问题”。它被其开发人员描述为“独特的、适应性强的区块链系统”和“世界上唯一可访问的神经形态量子计算云”。它利用基于 GPU 的挖矿网络的并行处理能力。

A neuromorphic network, Mapp shares with CoinGeek Backstage, would process information along the same principles as the human brain.

Mapp 与 CoinGeek Backstage 共享的神经形态网络将按照与人脑相同的原理处理信息。

“A neuromorphic network, in the simplest terms, is emulating the same way the human brain works. So when you’re looking at me, you’re seeing me, you’re hearing something around, you’re smelling—that is all parallel processing in your own brain. And we can simulate the same way that works on our chain,” he says.

“用最简单的术语来说，神经形态网络正在模仿人类大脑的工作方式。因此，当你看着我时，你正在看到我，你正在听到周围的声音，你正在闻到——这些都是你自己大脑中的并行处理。我们可以模拟在我们的链上运行的相同方式，”他说。

Dynex processes transactions using a custom algorithm called “DynexSolve,” which combines regular proof-of-work (PoW) and a new concept Mapp calls the “proof of useful work” (PoUW) mechanism. The idea of PoUW is to make sure a portion of the computational resources of Dynex’s mining network are dedicated to solving practical problems. Those would include real-world scenarios in scientific research, AI training, pharmaceuticals, logistics, and financial modeling.

Dynex 使用名为“DynexSolve”的自定义算法处理交易，该算法结合了常规工作量证明 (PoW) 和 Mapp 称为“有用工作量证明”(PoUW) 机制的新概念。 PoUW 的想法是确保 Dynex 挖矿网络的一部分计算资源专用于解决实际问题。这些将包括科学研究、人工智能培训、制药、物流和金融建模中的现实场景。

Its native token is also called Dynex or DNX. Miners receive DNX as a block reward for securing the network just as they do on other PoW blockchains, but also receive an additional subsidy for the “meaningful” computation described above. Mapp says that DynexSolve is a way of combining the two into one processing algorithm.

其原生代币也称为 Dynex 或 DNX。矿工会像在其他 PoW 区块链上一样获得 DNX 作为保护网络的区块奖励，而且还会因上述“有意义”的计算而获得额外补贴。 Mapp 表示，DynexSolve 是一种将两者结合为一个处理算法的方法。

He adds that one pharmaceutical research application Dynex is looking at involves protein folding. A pharmaceutical company noticed they’d previously done RNA folding and RNA sequencing.

他补充说，Dynex 正在研究的一项药物研究应用涉及蛋白质折叠。一家制药公司注意到他们之前做过 RNA 折叠和 RNA 测序。

“I think we’re up to around 250 amino acid chains. The human body has around 300 in a protein string, so we’re almost fully at human level of protein folding right now. We suspect a couple more months, and we’ll have that fully nailed, and that’s going to open the door to all kinds of new research,” he explains.

“我认为我们有大约 250 条氨基酸链。人体的蛋白质链中约有 300 个，因此我们现在几乎完全达到人类水平的蛋白质折叠。我们怀疑再过几个月，我们就会完全解决这个问题，这将为各种新研究打开大门，”他解释道。

LLMs (large language models) are a popular topic in technology, especially with ChatGPT and other similar projects gaining mainstream attention rapidly. Dynex LLM is the project’s home-grown model, and Mapp describes it as a “large behavioral model” based more on how a human mind thinks and learns.

LLM（大型语言模型）是技术领域的热门话题，尤其是 ChatGPT 和其他类似项目迅速获得主流关注。 Dynex LLM 是该项目的本土模型，Mapp 将其描述为更多基于人类思维和学习方式的“大型行为模型”。

A lot of Dynex’s work is still in the experimental realm, and its neuromorphic quantum computing concept utilizes ion drifting of electrons, which is different from superconducting qubit-based quantum computing in that it uses “memresistive elements that can quickly react to changes, helping the system swiftly find the best solutions.” It’s an alternative way to potentially achieve quantum computing advantages using the movement of ions within memresistors to change a device’s state.

Dynex 的很多工作仍处于实验领域，其神经拟态量子计算概念利用了电子的离子漂移，这与基于超导量子位的量子计算不同，它使用“可以快速对变化做出反应的忆阻元件，帮助系统迅速找到最佳解决方案。”这是利用忆阻器内离子的移动来改变设备状态来潜在实现量子计算优势的另一种方法。

This is just one computer science problem Dynex hopes to solve, and the team hopes for more breakthroughs.

这只是 Dynex 希望解决的一个计算机科学问题，团队希望能取得更多突破。