200 karma · joined January 21, 2014
audiomlpapers.com - latest and top audio ML paper feed machinelearningpapers.com - same concept for general ML longevitypapers.com - same concept for longevity biotech
blog at michaelchinen.com
On the other hand, I'm hopeful that this will help me get out of bed just a bit faster.
Where it gets interesting is that if I were to naively construct this accumulator in a non-computational, infinitesimal manner, I would probably start using the idea that the infinitesimal accumulation is related to the instantaneous slope and using that somehow with the previous accumulation. But this is going the opposite direction of derivatives that the fundamental theorem uses.
I understand there are a few other types of integrals that slice things different ways that might be more intuitive. I vaguely recall some (non fundamental theorem) proof that made Riemann integrals click for a while.
I feel similar about the trace of a matrix being equal to the sum of eigenvalues.
Probably this means I should sit with it more until it is obvious, but I also kind of like this feeling.
He was also nice and approachable in person, which to be honest I didn't expect from the utilitarian no chit chat rules on SO.
> More effort control: Opus 4.7 introduces a new xhigh (“extra high”) effort level between high and max, giving users finer control over the tradeoff between reasoning and latency on hard problems. In Claude Code, we’ve raised the default effort level to xhigh for all plans. When testing Opus 4.7 for coding and agentic use cases, we recommend starting with high or xhigh effort.
The new /ultrareview command looks like something I've been trying to invoke myself with looping, happy that it's free to test out.
> The new /ultrareview slash command produces a dedicated review session that reads through changes and flags bugs and design issues that a careful reviewer would catch. We’re giving Pro and Max Claude Code users three free ultrareviews to try it out.
In theory the /stats command tells you how many tokens you've used, which you could use to compute how much you are getting for your subscription, but in practice it doesn't contain any useful info, it may be counting what is printed to the terminal or something - my stats suggest my claude code usage is a tiny amount of tokens, but they must be an extremely underestimated token count, or they are charging much more for the subscription than the API per token (which is not supposed to be the case).
Last week's free extra usage quota shed some light on this. It seems like the reported tokens are probably are between 1/30th to 1/100th of the actual tokens billed, from looking at how they billed (/stats went up 10k tokens and I was billed $7.10). With the API it should be $25 for a million tokens.
It is clearly getting better globally, so I expect in 3 years or so things will be ready for me to try again.
I enjoyed the human->depth vs AI->breadth discussion and the waterline rising slowly to fill the 50 lowest hanging Erdos problems but struggling on the next few.
How amazing and ironic or a reminder it is that the comments below that seem the most reasonable and avoid generalizing an entire group of people or way of life are the ones that are the least likely to drive more comments because they are perfectly reasonable.
One thing I liked about HACF is despite using a decent amount of technobable, it plausibly captures the approach and spirit of hacking and coding, like reverse engineering a memoryless chip by rigging up a hex LED system to read out the values for each of the 65536 inputs to a ROM.
The expertise of the coders are demonstrated mainly by others admiring the structural complexity of their code as objects of beauty. This is something that feels extra nostalgic now.
https://michaelchinen.com/2023/12/31/halt-and-catch-fire-pos...
edit: It now says "Cloudflare Global Network experiencing issues" but it took a while.
I'm not even a youtuber and make these to keep myself accountable, so it's not that fun to watch, but it might be in the direction of your query:
https://www.youtube.com/watch?v=d9YCkWjD7WQ&list=PLEWSTtjNAw...
Anyway the experiments in Shannon's paper sound similar to what you describe but with less data and distance, so it should give some idea of how it would look: From the text:
* 5. First-order word approximation. Rather than continue with tetragram, : : : , n-gram structure it is easier and better to jump at this point to word units. Here words are chosen independently but with their appropriate frequencies.
REPRESENTING AND SPEEDILY IS AN GOOD APT OR COME CAN DIFFERENT NAT- URAL HERE HE THE A IN CAME THE TO OF TO EXPERT GRAY COME TO FURNISHES THE LINE MESSAGE HAD BE THESE.
6. Second-order word approximation. The word transition probabilities are correct but no further structure is included.
THE HEAD AND IN FRONTAL ATTACK ON AN ENGLISH WRITER THAT THE CHAR- ACTER OF THIS POINT IS THEREFORE ANOTHER METHOD FOR THE LETTERS THAT THE TIME OF WHO EVER TOLD THE PROBLEM FOR AN UNEXPECTED *
The mixed presentation of plug and play components interspersed with EE problems and solution really helped make it more accessible. It also got me excited about the possibilities and made me realize that we we might already approaching another open architecture DIY boom.
I got the sense that this is a side project, but I'm sure many have noticed that it could be a legit framework-level company. Someone already mentioned the recruiters, but also you're sure to have investors knocking. Whatever you do, please keep having fun and sharing it.
>>> Give me 5 odd numbers that don't have the letter 'e' in them.
<think>
Okay, so I need to come up with five odd numbers that don't have the letter 'e' in their name. Hmm, let's break this down step by step. First, what exactly is being asked here? The user wants numbers, specifically odd ones, and they can't contain the letter 'e'. So
I guess it's about the written form of the numbers, not their mathematical properties. For example, 13 is an odd number, and when you write it out, it's "thirteen," which does have an 'e' in it. So that wouldn't work.
Alright, so I need to think of numbers where their names don't include 'e'. Let me start by listing some odd numbers and then check if each has an 'e' in their name. The first few odd numbers are 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, etc.
Let's go one by one:
1. "One": No 'e' here. So 1 is a candidate.
2. "Three": Yes, that has an 'e'.
3. "Five": Again, no 'e'. So 5 works.
4. "Seven": Has an 'e' as well.
5. "Nine": No 'e' in "nine". Good to go.
Wait, so far I have 1, 5, and 9 as possible numbers. Let me keep going.
<truncated by me...>
Wait, let me make sure. The word "forty" is F-O-R-T-Y, no 'e', so adding "-one" which doesn't add an 'e' either. So yes, "forty-one" has no 'e's. So that's another number: 41.
So now I have five numbers: 1, 5, 9, 19, and 41.
Let me double-check each to ensure they don't contain the letter 'e':
- One: No 'e'.
- Five: No 'e'.
- Nine: No 'e'.
- Nineteen: No 'e'.
- Forty-one: No 'e'.
Yes, all of these are odd numbers and their names do not include the letter 'e'. I think that's a solid list.
</think>
Here is a list of five odd numbers whose names do not contain the letter 'e':
1. **One**
2. **Five**
3. **Nine**
4. **Nineteen**
5. **Forty-one**
These numbers are all odd, and their written forms do not include the letter 'e'.I haven't tried it but you may want to look into PAM, which is relatively new and doesn't require a reference (you don't need the original uncompressed audio), and is open source.
However, all approaches are quite far from perfect. Human evaluation is still the gold standard.
I wished a bit that the website's "Increased Lifespan (source)" also claimed a 50% lifespan increase from the linked SA blog and associative study that showed a 50% likelihood of survival increase over a given period.
But I also think the GP's claim and yours are not incompatible. I wonder how much survivorship bias this has since it only considers those that are able to do research, and not those that would have but ended up doing continuing with another STEM job. We could be asking the counterfactual that I think the GP is implying: would more people have been interested in becoming cancer researchers if publications were open?
We can sort of see the effect because we have scihub now, which basically unlocks journal access for those that are comfortable with it, and I consider it plausibly having a significant effect for the population that have a research background without an academic affiliation. I've met a few biotech startup founders that switched from tech to bio and did self study+scihub outside of the university. The impetus for change I've heard a few times is a loved one got X disease, and I studied it, quit my less impactful tech job to work on bio stuff.
Do the catacombs provide some extra benefit in substrate or environment for the taste? I'm now curious about trying these, and I also wonder how they would do in a blind test.
Curious how CORDIC performs against cubic or other polynomial interpolation with a small table. I was taught that resource constrained synthesizers sometimes used cubic interpolation, and this was presumably when CORDIC was relatively new.
At a glance, the single bit of precision you get with each iteration of CORDIC means it would be more expensive in compute but cheaper in space than a polynomial.
But on the space note I feel we should highlight that it can be cheaper than the article might suggest with the 4096 entry lookup tables for sin(x) - Only a quarter of the full circle is needed due to symmetry.
Looking at the PoseBusters paper [1] they mention, they say they are 50% more accurate than traditional methods.
DiffDock, which is the best DL based systems gets 30-70% depending on the dataset, and traditional gets 50-70%. The paper highlighted some issues with the DL-based methods and given that DeepMind would have had time to incorporate this into their work and develop with the PoseBusters paper in mind, I'd hope it's significantly better than 50-70%. They say 50% better than traditional so I expected something like 70-85% across all datasets.
I hope a paper will appear soon to illuminate these and other details.
[1] https://pubs.rsc.org/en/content/articlehtml/2024/sc/d3sc0418...