HNHacker News
TopNewBestAskShowJobs

mchinen

200 karma · joined January 21, 2014

Healthspan Capital (Longevity Biotech Fund) Manager Audio ML researcher/SWE ex-Audacity team member

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

submissionscomments
mchinen··on Virtual violin produces realistic sounds
Bowed instruments are very cool to model because of the nonlinear slip of the bow against the string. A bit curious why bowing was not discussed or used in the example of a violin, just plucking. Do luthiers test violins more by plucking than bowing?
mchinen··on Reddit seems to be blocking mobile browser access for some users
This a more aggressive form, the app redirect existed for a while, but there was the option to dismiss and continue using the browser.

On the other hand, I'm hopeful that this will help me get out of bed just a bit faster.

mchinen··on Fundamental Theorem of Calculus
Thank you for wording it better than I could. On the computational side, the integral as a function that measures how the another function accumulates intuitively lets you measure the area under the curve between two points only requires two variables.

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.

mchinen··on Fundamental Theorem of Calculus
I've studied the proofs before but there's still something mystical and unintuitive for me about the area under an entire curve being related to the derivative at only two points, especially for wobbly non monotonic functions.

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.

mchinen··on Thank You for Being a Friend
Jeff is a cool guy to make a reasonable and level headed post like this. I would be struggling to not be more resentful.

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.

mchinen··on Claude Opus 4.7
Does it run for you? I can select it this way but it says 'There's an issue with the selected model (claude-opus-4-7). It may not exist or you may not have access to it. Run /model to pick a different model.'
mchinen··on Claude Opus 4.7
These stuck out as promising things to try. It looks like xhigh on 4.7 scores significantly higher on the internal coding benchmark (71% vs 54%, though unclear what that is exactly)

> 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.

mchinen··on Pro Max 5x quota exhausted in 1.5 hours despite moderate usage
I've been feeling the squeeze too. I've tried switching between different models as a test, I can at least say it feels like the limits are about half of what they used to be a few months ago. I'd be totally willing to concede that this is just my perception if Anthropic would only release some tools for measuring your usage.

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.

mchinen··on Steam on Linux Use Skyrocketed Above 5% in March
This is similar to my experience. I find many people on the forums that can't resolve the crashes in Red Dead Redemption 2, so I suspect a lot of it depends on the specific games you've picked.

It is clearly getting better globally, so I expect in 3 years or so things will be ready for me to try again.

mchinen··on Mathematical methods and human thought in the age of AI
Thanks for breaking it down in a balanced way.
mchinen··on Mathematical methods and human thought in the age of AI
He also was on Dwarkesh's podcast last week (https://www.youtube.com/watch?v=Q8Fkpi18QXU ).

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.

mchinen··on Mathematical methods and human thought in the age of AI
I haven't seen any negative sentiment toward Terence Tao before. Coming from outside the academic math sphere, genuinely curious if there's a real issue or if this comment is just spam itself.
mchinen··on The happiest I've ever been
This has clearly triggered a lot of people, with the full spectrum of arguments for and against having kids below. I'm bookmarking this as an example of driving engagement by taking a fairly benign topic ('helping others was rewarding!') with an extreme view on a topic that everyone has a stake in ('an entire generation lost the plot on life').

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.

mchinen··on Halt and Catch Fire: TV’s best drama you’ve probably never heard of (2021)
I wrote a bit about how the kind of tech culture in HACF feels more relevant in light of the LLMs even 2 years back before I heard of mourning a craft. Here's an excerpt:

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...

mchinen··on Size of Life
If you liked the first half of this site and want an extension, Cell Biology by the Numbers (2015, Milo, Phillips, https://book.bionumbers.org/) is great and has a similar intuition-building fun sense about size as well as various other measurements, including weight, time and energy at the atomic to micro-organism level.
mchinen··on Cloudflare Global Network experiencing issues
Down, but the linked status page shows mostly operational, except for "Support Portal Availability Issues" and planned maintenance. Since it was linked, I'm curious if others see differently.

edit: It now says "Cloudflare Global Network experiencing issues" but it took a while.

mchinen··on Why do we remember some life moments but not others?
Definitely! This makes me think of petrichor, the earthy smells released during rain, in urban environments is very local and memory triggering because it's tied to bacterial soil/rock/plant composition. There's something visceral about it for me because my default model of memories is audio and video playback, and the smell hits me with a forgotten dimension when I go back to a place and it rains.
mchinen··on ChatGPT Pulse
Scott Alexander has a similar 1-page horror story from 2012 where an earing that gives advice consumes the user: https://gwern.net/doc/fiction/science-fiction/2012-10-03-yva...
mchinen··on Vibe coding tips and tricks
I recorded myself trying it out to port some old apps of mine using Claude code as a first time user of it.

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...

mchinen··on Deep Learning Is Not So Mysterious or Different
Claude Shannon was interested in this kind of thing and had a paper on the entropy per letter or word of English. He also has a section in his famous "A Mathematical Theory of Communication that has experiments using the conditional probability of the next word based on the previous n=1,2 words from a few books. I wonder if the conditional entropy approaches zero as n increases assuming ergodicity. But the number of entries in the conditional probability table blows up exponentially. The trick of combining multiple n=1 of different distances sounds interesting, and reminds me a bit of contrastive prediction ml methods.

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 *

mchinen··on Show HN: I made an open-source laptop from scratch
This is really one of the best things I've seen on HN in 15 years.

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.

mchinen··on DeepSeek-R1
[Edit: I suppose the correct answer was for the full model]. What I got was much worse when I ran with hf.co/unsloth/DeepSeek-R1-Distill-Llama-8B-GGUF:Q8_0. (I think the glyph-based reasoning is especially hard for models though, since they have a token-level blindness to them):

  >>> 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'.
mchinen··on Ask HN: Must-Read Books for Startups?
All the 'founder attitude' book recs are probably a good idea, but I would recommend supplementing with some VC and investing vehicle stuff, such as Secrets of Sand Hill Road and The Power Law, especially if that seems unimportant and tedious. Understanding your options and what the incentives are for investors will help you better understand where you are likely to be aligned and where you have to be careful.
mchinen··on Ask HN: Determine sound quality of a file, objectively?
Most of the work on objective quality metrics (e.g. PESQ, POLQA, ViSQOL, DNS-MOS, NISQA) focus on speech because of telecommunications demands, but some of these have an audio mode. But there are some new promising audio ones that are ML based.

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.

mchinen··on In ‘The Book Against Death,’ Elias Canetti rants against mortality
I'm also ex-Google working in the space on the funding and company formation side. Would love to chat.
mchinen··on Show HN: Rent Elderly People – Skip Lines and Get Senior Perks
I enjoyed this.

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.

mchinen··on AI's $600B Question
I also think the bottleneck isn't access to the papers today and data access and silos are more important.

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.

mchinen··on Paris's Catacomb Mushrooms (2017)
I'm assuming the mushroom in question is agaricus biosporus in white button form. Interesting to hear the 3 star Michelin chef say there is a huge difference between paris mushrooms and industrial mushrooms of the same species due to the natural way it grows, because mushroom foragers often find this species in the wild and it's cousin agaricus campestris, but I never hear anyone mention it is particularly better than store bought mushrooms. The wild ones I've tried tasted fairly similar. For other cultivated mushrooms like lion's mane or oyster, foragers also don't claim huge taste benefits to the wild ones, and there is usually some fly larvae even if they look good.

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.

mchinen··on Why the CORDIC algorithm lives rent-free in my head
Nice to read this.

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.

mchinen··on AlphaFold 3 predicts the structure and interactions of life's molecules
I am trying to understand how accurate the docking predictions are.

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...

← PreviousPage 2 of 3Next →