[dead]
Untitled topic
31–40 of 61 posts
Re: undefined
#32So, if I got this right, this is just about re-implementing an existing load balancing algorithm faster...? If so, this is really dumb. As you guys checked out, yes most load balancing algorithms are slow/dumb: >First, we evaluate DeepSeek's open-source EPLB implementation. This employs a greedy bin-packing strategy: experts are sorted by load in descending order, and each is placed onto the least-loaded GPU that has…
Agree. Starting from Python for-loops is embarrassing baseline. Any decent implementation gets you most of that 5x for free. The interesting part isn't the speedup - it's that AI can do routine optimization unsupervised. That's the actual value prop.
Re: undefined
#33The idea that AI can discover anything is ridiculous. It can propose algorithms like it creates any piece of text, but only the human researcher is capable of analyzing the algorithm, proving that it works, understand what it is doing, i.e., pretty much everything that we call a new "discovery". I would have zero confidence in an algorithm "discovered" by an AI in isolation.
Re: undefined
#34There's nowhere on the page to find out what "ADRS" stands for since the upper left is cut off and isn't a link to your home page.
Re: undefined
#35I'm not sure if this is the exact same thing, but a load balancing paper reported a 4.2x speedup by applying a "snake pattern" in 2024: https://arxiv.org/pdf/2402.02447
Most probably the AI was secretly tested on this data and is just stealing the algorithm.
Re: undefined
#36So, if I got this right, this is just about re-implementing an existing load balancing algorithm faster...? If so, this is really dumb. As you guys checked out, yes most load balancing algorithms are slow/dumb: >First, we evaluate DeepSeek's open-source EPLB implementation. This employs a greedy bin-packing strategy: experts are sorted by load in descending order, and each is placed onto the least-loaded GPU that has…
Thanks for commenting! Actually in this case, "the work being done" can be really fast because it can be done asynchronously. For context, here’s how this translates in a real-world application.
The original algorithm was provided by DeepSeek, and our optimized implementation achieves a 92× speedup over it. The 5x number is comparing with another baseline that is undisclosed yet.
When integrating EPLB into vLLM, I discovered—somewhat unexpectedly—that the open-source algorithm consumes nearly half of the total time of a rearrangement step, with the remaining time spent transferring weights across GPUs. To address this, I applied OpenEvolve to the algorithm, setting the primary objective to improve speed while maintaining the same balance factor. It performed remarkably well. With additional optimizations on the weight transferring, the overall overhead has now become almost negligible.
Re: undefined
#37The idea that AI can discover anything is ridiculous. It can propose algorithms like it creates any piece of text, but only the human researcher is capable of analyzing the algorithm, proving that it works, understand what it is doing, i.e., pretty much everything that we call a new "discovery". I would have zero confidence in an algorithm "discovered" by an AI in isolation.
Theoretically if I were to type into an LLM "Write a novel compression algorithm for images that is at least 25% smaller at the same speed and quality as ___" and it did, and I ran the code (which I didn't understand) and it worked, wouldn't that count?
The odds of that working, though, are of course pretty near 0. But theoretically, it could happen.
Re: undefined
#38The idea that AI can discover anything is ridiculous. It can propose algorithms like it creates any piece of text, but only the human researcher is capable of analyzing the algorithm, proving that it works, understand what it is doing, i.e., pretty much everything that we call a new "discovery". I would have zero confidence in an algorithm "discovered" by an AI in isolation.
Theoretically if I were to type into an LLM " Write a novel compression algorithm for images that is at least 25% smaller at the same speed and quality as ___ " and it did, and I ran the code (which I didn't understand) and it worked, wouldn't that count? The odds of that working, though, are of course pretty near 0. But theoretically, it could happen.
You might find that if it did produce something, it might not be _novel_
Re: undefined
#39The idea that AI can discover anything is ridiculous. It can propose algorithms like it creates any piece of text, but only the human researcher is capable of analyzing the algorithm, proving that it works, understand what it is doing, i.e., pretty much everything that we call a new "discovery". I would have zero confidence in an algorithm "discovered" by an AI in isolation.
Theoretically if I were to type into an LLM " Write a novel compression algorithm for images that is at least 25% smaller at the same speed and quality as ___ " and it did, and I ran the code (which I didn't understand) and it worked, wouldn't that count? The odds of that working, though, are of course pretty near 0. But theoretically, it could happen.
As you say, the odds of this happening are very close to zero. But suppose for a minute that this was possible. Did you learn anything? Do you really have a discovery? Was this done using a novel method or applying something that already exists? If you give this to somebody else, should they believe it works? Is this result even understandable by human beings? You'd need to answer so many questions that in the end even this would NOT be a discovery by the machine but by yourself.
Re: undefined
#40The idea that AI can discover anything is ridiculous. It can propose algorithms like it creates any piece of text, but only the human researcher is capable of analyzing the algorithm, proving that it works, understand what it is doing, i.e., pretty much everything that we call a new "discovery". I would have zero confidence in an algorithm "discovered" by an AI in isolation.
It can propose algorithms which than it can _itself test and iterate on_