正在加载视频...

视频加载失败

37,014 次观看 • 1 个月前 •via X (Twitter)

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Andrew Ng just revealed why the AI companies throwing the most compute at the problem are going to lose. The winner of the intelligence race won’t use the most compute. They’ll waste the least. Ng: “Most of your high-dimensional data lies on a lower-dimensional subspace. It’s just a fact of life.” Here’s what that means in practice. You have a 10,000-dimensional dataset. Every dimension dragged through every calculation. Every training cycle hauling dead weight the model will never use. Ng: “You’re carrying around these 10,000-dimensional examples throughout your whole training process.” That bloat isn’t just inefficient. It’s a tax on every computation you run. Memory bandwidth. Network bandwidth. Computational speed. All of it eaten by dimensions that contribute nothing to intelligence. They contribute noise. The insight that separates the architects from the arms race: that 10,000-dimensional dataset is almost entirely captured by a much smaller subspace. The signal lives in a fraction of the space you’re paying to process. Compress it. 10,000 dimensions down to 1,000. Ng: “You can run your learning algorithm on a much lower-dimensional set of data and it may be much more efficient.” Same hardware. Same budget. A fraction of the friction. Brute force is the strategy of whoever has the deepest pockets. Compression is the strategy of whoever actually understands the problem. The companies that master this don’t just build faster models. They build models that find more truth in less data than anything scaling blindly ever will. Intelligence was never about processing everything. It’s about knowing what to cut.

Dustin

215,287 次观看 • 5 个月前

Super Shrimp World Level 6 - Tokenomics! Check out the Galxe page for level 6 here: This quest attempts to cover the last missing pieces of information around SHRIMP, mainly our tokenomics. Tokenomics are weird and complicated, and can be difficult to understand from a regular users point of view - we'll try to make it as easy to digest as possible, by providing proper context and transparency. Our tokenomics are designed by us, which means they might be flawed, but they are carefully chosen this way because we believe it's the best course of action considering our ambitions for SHRIMP. There will be a fancy pie-chart on our website, but to summarize, here is what it looks like: Total supply: 121,212,121. 30% - Community. 30% - Reserve. 15% - LP. 15% - Investors. 5% - DAO. 5% - Team. Here is a brief description for each category: Community: Community is sort of a broad term - this allocation includes, but is not limited to: Aptomingos, [REDACTED] Keys and CB ordinals: Our dear loyal supporters - the people who have faith in us. Aptomingos will naturally get the highest allocation, but we'd like to show good faith to both key holders and CB ordinal holders by providing a small allocation to them as well. SHRIMP liquidity providers: We'd like to incentivize you guys to provide liquidity for SHRIMP, and we'll be working closely with DEX's to make this a reality. Barbecue participants: The Barbecue is all about rewarding active users who enthusiastically pays attention to our shenanigans, and punish the ones that don't. NFT and DeFi communities: The SHRIMP experience isn't gated to our holders only - we're looking to include the entire Aptos ecosystem, and even a few friends from other networks. This means we'll drop an allocation to our NFT and DeFi friends as well. This won't be done in a regular fashion, but we'll provide more context on this front in the near future. Galxe participants: These Galxe campaigns have been part of a little way to boost our engagement while still educating you guys about the core concepts of SHRIMP. So far, they've been pretty easy, but you never know what might happen in the future. Reserve: This allocation of tokens are specifically set aside for our mechanic called The Reserve. These tokens will only be utilized when we hit certain burn thresholds from The Barbecue. We highly doubt that we'll utilize the total amount in this category this way, so this allocation will also serve as a backup for things like grants for builders, more airdrops, providing more liquidity, etc. LP: Our initial liquidity pool will consist of 15% of the total supply. We'd like for people to able to trade freely, and sufficient liquidity is the key to that goal. Investors: We believe we have a good mix of people in our presale round. We've got founders, real products, influential traders, whales, and people who've shown loyalty and faith in us from our humble beginnings. This allocation is vested, and they will have ZERO tokens available to trade on day 1. DAO & Team: The DAO allocation is for anything we'd like to do as a project. Think of this allocation as a jack-of-all-trades. Team allocation is sort of self-explanatory, we'd like to be involved in the process and experience of SHRIMP, while also being paid for the work we've done over the past couple of months. Team allocation is vested, similar to presale. We retain the right to drop these allocations as we see fit, and we've already explained that not everything will be dropped at once. We'd like a healthy circulating supply, which means things might be done slowly, and might be adjusted as we go.

SHRIMP

129,341 次观看 • 2 年前