waldrews

9 ms·

91.
by waldrews 2y ago
Repeal the Second Law! Free Energy for Everyone!
92.
by waldrews 2y ago
It's a classic application of the pigeonhole principle, the first on in this list: https://en.wikipedia.org/wiki/Pigeonhole_principle#Uses_and_...
93.
by waldrews 2y ago
No invertible function can map every non-negative integer to a lower or equal non-negative integer (no perfect compression), but you can have functions that compress everything we care about at the cost of increasing the size of things we d
94.
by waldrews 2y ago
It would average the same size as the actual data. Treating the pi bit sequence as random bits, and ignoring overlap effects, the probability that a given n bit sequence is the one you want is 1/2^n, so you need to try on average 2^n
95.
by waldrews 2y ago
Why go through all that effort, when it's just tau/2.
96.
by waldrews 2y ago
For those of us who were last paid to write C++ in a different millenium, suggestions for a book to catch up?
97.
by waldrews 2y ago
The way I've been telling it: - It's statistics if the goal is to figure out why it works - It's data science if nobody cares why it works - It's ML if nobody knows why it works - It's AI if it's actually impos
98.
by waldrews 2y ago
Yes! Now, we just have to run this inside a browser, which will run inside a container, which will run inside a VM, which will run on an emulation layer...
99.
by waldrews 2y ago
Feels like one of those technologies where you read about it, and it seems cool/powerful and you wish you had a use case to try it out, but can never find one that you can't handle more easily without it. Am I missing some big cu
100.
by waldrews 2y ago
Is there an out-of-the-box solution that just does all of this - as a hosted solution, a boilerplate, or a best-practices example? Maybe Supabase could do it? Would be nice to one-click deploy a Postgres with cache and auditing and the re
101.
by waldrews 2y ago
But temperature 0 LLM's don't exhibit the emergent phenomena we like, even in apparently non-creative tasks. The randomness is, in some sense, a cheap proxy for an infeasible search over all completion sequences, much like simula
102.
by waldrews 2y ago
This seems like a good place to ask - if your knowledge of (computation/complexity) theory is at the level of the Sipser Introduction to the Theory of Computation book, what's a good learning path to get current?
103.
by waldrews 2y ago
What's the deal with hosted open source model services not supporting grammars? I've seen fireworks.ai do it, but not anybody else - am I missing something?
104.
by waldrews 2y ago
Ironically, .Net has several.
105.
by waldrews 2y ago
Hetzner GPU servers at $200/month for an RTX 4000 with 20GB seem competitive. Anyone have experience with what kind of token throughput you could get with that?
106.
by waldrews 2y ago
do you mean 1.28 TB?
107.
by waldrews 2y ago
The Clojure/JVM statistics/scientific computing/now tensor math packages just never got as good as Python, and in Smalltalk they were a non-starter. R is an awkward language, but it's repl-first, has a lot of Lisp'
108.
by waldrews 2y ago
Notebooks are a diluted form of the Lisp/Smalltalk REPL-based development experience, with some features from the reactive-spreadsheet world. Especially for those of us in the business of producing numbers, 'real code' with
109.
by waldrews 2y ago
Would we even hate the ads so much if they so many of them weren't overtly scammy? Give us quality ads and have a working system for reporting scammy or otherwise inappropriate ads, and the resistance will be much lower. A premium
110.
by waldrews 2y ago
As previously done by the rock group KMFDM.
111.
by waldrews 2y ago
Hooray for containerized models, but a game changing feature for such containers could be K-V/prefix caching. Previous discussion: https://news.ycombinator.com/item?id=40034972#40036309
112.
by waldrews 2y ago
Is the best match dot product (or other distance metric used) below some level? What the good level is would have to be examined for your particular use case, maybe make a histogram of the best (or several best) dot products from many good
113.
by waldrews 2y ago
Clojure, ClojureScript, and LFE are all production-ready, primarily because they have straightforward integration with existing VM's/ecosystems (JVM, JS, and BEAM respectively). Sadly the .Net Clojure seems to be unmaintained, and
114.
by waldrews 2y ago
Anyone who has spent enough time in AR to compare, how's the headache/eye fatigue compared with VR?
115.
by waldrews 2y ago
They promise free (domain) email hosting, that's kind of interesting... Anybody see a catch (other than, it's ByteDance?)
116.
by waldrews 2y ago
It's the centenary of Kafka's death today. Remember him by doing something absurd and horrifying.
117.
by waldrews 2y ago
At something like a cent per message, both ways, SMS is annoyingly expensive for any application with a lot of back and forth traffic - hence, chatbot-over-SMS solutions are rare, and auth providers charge extra for SMS based verifications.
118.
by waldrews 2y ago
OpenAI's speech to text API is the official Whisper implementation. It's expensive. Many small providers will host Whisper as an endpoint (Replicate for example). Groq has Whisper in a private beta; if they open it up for commerc
119.
by waldrews 2y ago
Could you talk about your choices for speech thresholds/silence detection? GPT-4o supposedly can speak with interruptions, but for the rest, presumably you have to either do VAD or some ad-hoc volume based rules to detect start and st
120.
by waldrews 2y ago
You don't do it for free, but you also don't charge so much it becomes a profit center and screws up incentives.

More ›