howlin

One thing to keep in mind about direct (learn the policy/behavior) versus indirect (learn the model and then simulate behaviors on the model to choose the best) is that sometimes it's much easier to find a good enough policy than it is to learn an accurate enough model for simulation. Driving is a good example of this. Most of the time all you need to do is stay in your lane and obey the rules for intersections. A simulation of a driving environment, on the other hand, is quite difficult.

This is a truly terrible perspective. Many pedestrians are slow and extremely vulerable to the effects of being toppled by a wheeled vehicle. This is particularly a problem when people are leaving buildings onto sidewalks unaware of cross traffic moving at 10+ kph. Bicycles belong on the road with other vehicles.

There are several responses to the "plants have senses" rebuttal:

If you care about the well being of plants, you should still prefer to eat plants rather than animals. The trophic level of animals in the food chain necessitates that they will eat more calories of plant matter than would provide calories of food upon being eaten.

Secondly, you should look at the sorts of plants being consumed and where they fall in the plant's life cycle. Fruits and nuts have been evolved to be eaten by animals. Staple crops such as grains and pulses are generally harvested from plants that have already lived to maturity and died off. Really it's only fresh greens and tubers that require the damage or destruction of a living plant.

Lastly, it's important to recognize the evolutionary purpose of sensing pain and suffering from it. Animals use pain and suffering to detect harmful situations and learn how to avoid them in the future. This process consists of sensations, memory and adaptive behavior. Plants may have "reflexive" responses to stimulus, but these responses aren't adaptive and aren't affected by previous experience (mostly... there are some exceptions). There would be no evolutionary purpose for plants to develop a sense of suffering if they have no practical use for that sensation.

I'm pretty sure it's daughter-father, not wife-husband.

> If your neural net (haven't actually met anyone who uses these in trading) takes 5ms to make a prediction that'll lock you out of a whole lot of trading opportunities/strategies.

Let's say you have trading opportunities once every 100ms. They need to be acted upon within 2ms or they vanish.

You don't have the time budget to run a NNet every time the state of the market changes. You can, however, train a NNet to output a very small decision tree that can run in under 1ms. The NNet can then decide which "micro-strategy" in the form of a much faster and more reactive decision tree is more appropriate for the current market context.

> Basically, there's a bit of a spectrum of fast vs smart. In general it's hard to do incredibly smart stuff fast enough to compete in the "speed-critical" bucket of trades and vice-versa there's barely any point in being ultra-fast in the "non-speed-critical" bucket because your alphas last for minutes to hours.

Depending on the overall strategy, you can be smart and fast at once. If you have occasional alpha harvesting opportunities that must be acted on quickly (very low latency but also low-ish frequency), then it is possible to spend the time in between trades modeling market context and developing optimal short-term plans for how to react in case of various market triggers.

There are some rather archaic behaviors of the versions of these tools that come by default on MasOSX.

For instance, on a mac the 'sed' command doesn't understand the difference between space and tab

https://stackoverflow.com/questions/40311339/using-sed-on-ma...

Especially in the early days, the main competitors to online retail were catalogue services. I would think they have most business and logistics problems in common, minus running a web site.

I'd say a 2% mortality rate to ship chicks is unacceptable. We've practically shut down the entire world economy because of a 0.5-2% mortality rate from Covid-19.

It's a symptom of a bigger issue. These hatcheries even under the best of circumstances are creating lives that will be hard and short just to turn a profit.

Yes, I am very well aware of this. Which is why I don't eat chickens, other animals, or consume any other livestock product.

But every once in a while I forget the fractal nature of the horrors we inflict on any animal that is unlucky enough to get caught up in our agricultural industry.

From the article:

Usually they arrive every three weeks like clockwork,” she said Wednesday. “And out of 100 birds you may have one or two that die in shipping.

The wrongness of what's happening here is not just that the Post Office is experiencing unusual delays. Even the normal state of affairs shows a grotesque indifference to life.

Soy milk is just as reasonable environmentally and probably a little healthier as a nutrition source compared to oat milk. Though the flavor of soy is a little harder to get used to.

The 1950's were an interesting time for grand "mad science" engineering projects. Anyone in the Bay area should make an effort to see the Bay Model museum in Sausalito. It contains a huge model of the SF Bay and the local watershed. While this is interesting, it is more interesting to hear the story of why it was made. It was intended to vet a plan to dam the bay to turn it from salt water to a huge freshwater lake. Fortunately they decided it wasn't feasible.

Which is a great reason to eat as low on the food chain as you can.

There are still frontier nations out there. Russia, Greenland, Outback Australia, the Canadian Territories. Perhaps Alaska.

But the continental US is a developed land with dense cities. Your idea of what your liberties should be don't justify the drastic externalities you will inflict on your neighbors.

He's been working on Go for a very long time. Since he was a PhD student. Though I am sure the team helped, don't think he wasn't the main driving influence behind the algorithms and featurization of the problem.

You'd need a matching property in the southern hemisphere for the dark winters.

It's best to follow the SP500 rather than the Dow. The former is much more representative of the whole market. And it's best to follow market movement in percentage rather than points. Points inflate over time as the nominal value of the market indexes climb.

It's pretty clear the crime here was unauthorized access to the server, not the sending of the information on to the media.

Palm oil effectively replaced partially hydrogenated oils in various junk foods when we learned how unhealthy trans fats were. Fully hydrogenating these oils will not have exactly the same properties as partial hydrogenation, but for many purposes it will work ok in many of these junk food applications. Fully hydrogenated oils don't have the trans fat issue, but it will require consumer education to make this point clear.

To be fair to the poster, virtually anyone you meet in real life will be far less interesting than chatting with someone in a specialized forum like hacker news.

It's more likely the salt than the wheat. From the article:

| The traditional Japanese diet was thought to increase stroke risk because of the heavy use of seasonings such as miso, salt, and soy sauce.

Bayesian modeling can be very powerful when it works but it can also be catastrophic when it fails. It helps to think about this in an adversarial decision theoretic context where you play a prediction game against an opponent (usually called Nature).

We can think of the game as discovering the best model to explain a set of observations. The Bayesian believes that Nature picks the true model that generated the observations by sampling the prior. This is actually a huge assumption to make, which is why Bayesian methods work so well when the assumption is close to the truth.

Frequentists make the assumption that Nature chooses the underlying true model from a set of possible models. Beyond restricting the set of models Nature can choose from, frequentists make no further assumptions about the selection process. This is a strictly weaker assumption than the Bayesian makes, which means frequentist methods will do better when the specified prior grossly misrepresents Nature's decision making process.

There are even weaker assumptions that can be made about how Nature chooses the data. Regret-based model inference allows for a more adversarial game with Nature where the data may not come from the class of models considered at all. If Nature truly behaves this way, then Bayesian decision making can catastrophically fail.

There are a lot of ways of formalizing the problem of recommendation. Perhaps the variant of the problem used by Netflix is "solved", but it's kind of an odd one. Basically, they built a system to answer questions of the following form: "Given that user X watched media Y, what rating would they give it?" They trained and tested on media that users have already rated. Some of the ratings are masked and thus need to be "predicted" for the test.

The issue is that the Netflix dataset has a baked-in assumption that a recommender system should show media that a user is likely to have ranked highly. It may be more important to show the user media they wouldn't have found (and thus ranked) at all. Or perhaps a user will be more engaged with something controversial rather than generically acceptable. Who knows?

The main problem here seems to be that the author was working from home. A tiny house is really only good for washing, sleeping and lounging.

"What makes us so confident that plants don't also meet this criteria?"

Brains or other possible organs for processing information are metabolically expensive to operate and entropically expensive to maintain the genetics required to build them. If there is no evolutionary advantage for plants to have sensations and emotions, they will not evolve the machinery required to have them.

Note that this is mostly about saturated fat sources versus polyunsaturated fats. These polyunsaturated fats were all high in omega 6, which is widely regarded as less heart healthy than omega 3 oils or monounsaturated fats such as olive oil, avocado oil, or high oleic sunflower oil. Equivocating the results here as plant fats versus animal fats misses most of the nuance.

I wouldn't accuse these neighborhood stores of price gouging as much as I'd just say they are pricing high enough to cover the much more expensive cost of leasing or owning the commercial space.

Literally all of the problems you mentioned existed before GMO was a thing. Companies have been patenting varietals since the 1930's. Patent trolling in agriculture is also quite old. E.g. the yellow bean patent debacle:

https://www.nature.com/news/2008/080507/full/453145b.html

GMO is tangentially related but few of the biggest overreaches in agricultural IP are directly tied to them.

You'd be taking on currency exchange risk, as the Danes will want to loan you Euros but you'll likely be paying back the loan in USDs.