Hacker News: highd

New comment by highd in "Pro Max 5x quota exhausted in 1.5 hours despite moderate usage"

highd — Sun, 12 Apr 2026 19:34:02 +0000

... so how do API users enable 1hr caching? I haven't found a setting anywhere.

New comment by highd in "High-Level Is the Goal"

highd — Sat, 17 Jan 2026 06:10:15 +0000

I'm fairly confident that the new reddit React implementation can be improved in performance by a factor of 3x to 10x. I would be interested to hear others who have good reason to explain why not. I can certainly imagine React-like systems that are capable of statically determining DOM influence sufficient to make comment-collapsing negligible.

New comment by highd in "I just trained a physics-based earthquake forecasting model on a $1000 GPU"

highd — Tue, 04 Nov 2025 06:47:47 +0000

How are you doing your train/test split?

New comment by highd in "U.S. agencies back banning top-selling TP-Link home routers on security grounds"

highd — Thu, 30 Oct 2025 20:29:45 +0000

If you think identifying a hypothetical back door in something as complex as a router is not a difficult process you are simply uninformed.

New comment by highd in "Galleri test: Exciting results from blood test for 50 cancers"

highd — Tue, 21 Oct 2025 04:35:35 +0000

More like classic Bayes. Taleb's pop-sci came a few decades later.

New comment by highd in "Show HN: I invented a new generative model and got accepted to ICLR"

highd — Fri, 10 Oct 2025 15:19:56 +0000

Do you have any details on the experiment procedures? E.g. hardware, training time, loss curves? It is difficult to confidently reproduce research without at least some of these details.

New comment by highd in "A Comprehensive Survey of Self-Evolving AI Agents [pdf]"

highd — Wed, 13 Aug 2025 22:09:51 +0000

If you released any of the software you are using to do this I would find it extremely interesting!

New comment by highd in "Q-learning is not yet scalable"

highd — Sun, 15 Jun 2025 18:27:35 +0000

You can do Q-Learning with a transformer. You simply define the state space as the observation sequence. This is in fact natural to do in partially observed settings. So your distinction does not make sense.

New comment by highd in "Q-learning is not yet scalable"

highd — Sun, 15 Jun 2025 17:05:42 +0000

TFP cites decision transformers. Just using a transformer does not bypass the credit assignment problem. Transformers are an architecture for solving sequence modeling problems, e.g. the credit assignment problem as arises in RL. There have been many other such architectures.

The hardness of the credit assignment problem is a statement about data sparsity. Architecture choices do not "bypass" it.

New comment by highd in "Deep Quaternion Networks (2017) [pdf]"

highd — Sat, 24 Mar 2018 05:49:29 +0000

Can you provide some technical details on what you do? Do you divide the number of channels on each real layered network by 4? I don't see anything describing this in the paper.

New comment by highd in "Rent out GPUs to AI researchers and make ~2x more than mining cryptocurrencies"

highd — Sat, 24 Mar 2018 04:12:10 +0000

Golem is solving exactly this problem.

New comment by highd in "Stock Price Prediction with LSTMs"

highd — Wed, 21 Mar 2018 21:07:38 +0000

It's actually using just one day's data (one point) to predict the next day's data, let alone predicting 5 days.

New comment by highd in "Stock Price Prediction with LSTMs"

highd — Wed, 21 Mar 2018 20:57:23 +0000

This model is showing predictions one day into the future. The "test set" plot is all predictions made with data from 1 day ago. The input sequences have size 1, so no recurrence is happening (see to_1dimension in https://github.com/miguelgfierro/sciblog_support/blob/master...). If you predict that tomorrow will have the same price as today you'll get better plots under the same operating conditions.

In[2]: TIME_AHEAD = 1

Train set has ~1e-6 MSE, Test set has ~0.8 (0.94^2).

EDIT: I should say this person is probably learning, and a lot of this is honest mistake.

New comment by highd in "Death of the sampling theorem?"

highd — Wed, 21 Mar 2018 04:19:40 +0000

Most of the research in compressed sensing type applications now is focused on more sophisticated prior distributions than sparse/Laplacian. These more general bayesian approaches provide significant performance improvements over LASSO etc.

New comment by highd in "Deep Quaternion Networks (2017) [pdf]"

highd — Sun, 18 Mar 2018 05:06:41 +0000

Not to mention that it appears they're comparing against networks of the same architecture. If you build your quaternion components with with same number types as your reals you effectively have 4 times the number of parameters, which could be most of the benefit. They should also benchmark against similar architectures with equivalent parameter counts.

New comment by highd in "Predicting outcomes for games of skill by redefining what it means to win"

highd — Tue, 06 Feb 2018 07:53:56 +0000

Regarding this and previous discussions on this topic on HN, it seems to me that one of the primary motivating factors when constructing a new ranking system should be the possibility of cyclical dominance a la rock/paper/scissors, which we should expect to see in many modern complex games. If one wanted to solve that problem, I think it would require some form of multidimensional ranking. Then one would have added flexibility so you could accurate predict winners from the scoring with some function f(x_1,x_2) even in cyclical cases. This would be less interpret-able, but it would have a fair shot at actually modelling/predicting/depicting the dominance relationships between players.

New comment by highd in "Show HN: A radically more scalable alternative to Bayesian Optimization"

highd — Sat, 27 Jan 2018 03:46:07 +0000

Bayesian optimization is often applied on problems where the cost of function evaluation greatly exceeds the cost of any posterior updates etc. In this case sample efficiency is the real benchmark, and you should be looking at realistic problems with d>>2.

New comment by highd in "The lucrative business of America’s opioid crisis"

highd — Thu, 28 Dec 2017 21:48:59 +0000

Investors that investigate companies, identify hidden problems and report on them to the public help enable more accurate price discovery. This wouldn't happen at all if they couldn't trade against it.

For instance, I recall a story where a hedge fund sent PIs to determine that a factory reported running was actually closed down. They can't just trade against that info - it also has to be made public for them to realize gains. I can't imagine how you'd consider activity like this "getting away with it", when the alternative is companies performing unscrupulous acts in secret.

New comment by highd in "MobileCoin: A New Cryptocurrency from Moxie Marlinspike"

highd — Sat, 16 Dec 2017 01:18:22 +0000

This is a great example of where crypto would be useful. But it's not the same as paying for pizza.

New comment by highd in "MobileCoin: A New Cryptocurrency from Moxie Marlinspike"

highd — Sat, 16 Dec 2017 00:47:06 +0000

Cash is cheaper for merchants, but people have only continued to use credit cards more. Further, if crypto becomes attractive from a cost perspective Visa will just be forced to bring down fees and stay competitive. It's pretty near impossible to imagine crypto with costs less than Visa's database.