“Statistics are like bikinis. What they reveal is suggestive, but what they conceal is vital.”
-Aaron Levenstein, associate professor of business administration at Baruch College from 1961 to 1981
I’ve also heard it as “Statistics are like bikinis. What they reveal is exciting, but what they conceal is critical.”
Game development is now as much art as science, or rather the art of science. Even something as simple as how and when to use randomness can profoundly impact the fun of a game. Enter the observation of two different kinds of randomness: input and output randomness.
Input randomness is randomness that is decided BEFORE a player makes their strategy and decisions. Examples would include having a random number of enemies generated before the fight starts. While the number is random, knowing how many will show up actually lets the user decide to use different strategies and feel more in control.
Output randomness is often a big contributing factor to frustrating parts of gameplay. Examples here would consist of attacking an enemy, only to find out your attack completely missed out of sheer bad luck or an usually bad hit roll. This kind of behavior, while mathematically correct, often leaves users feeling like they were ‘robbed’ and that the game is cheating.
Games are increasingly using input randomness as a way to give users control. Even games that rely on output randomness often put their thumbs on the scales so that you do not lose as often as you’d like. In Civilization, if your unit with a 33% chance of hitting misses twice in a row, it’s guaranteed to hit on the 3rd try – even though real randomness wouldn’t behave like that.
Anyway, this is a great video about the different kinds of randomness.
Just down the street from me is an old firehouse. It was retired years ago and now serves as a private residence. This last year for Halloween, they actually partly converted it to a Ghostbuster themed Vacasa complete with props and even a really cool replica ghostbuster ambulance. How’s that for cool?
Here’s an article and pictures of it renovated into a home:
Links:
It has not been a good century for deterministic views of the universe.
In the early 1900’s, Gödel destroyed the work of many prominent mathematicians by proving that no one mathematical model will ever be able to describe reality.
He proved that any set of axioms you could posit as a possible foundation for math will inevitably be incomplete; there will always be true facts about numbers that cannot be proved by those axioms. He also showed that no candidate set of axioms can ever prove its own consistency.
His incompleteness theorems meant there can be no mathematical theory of everything, no unification of what’s provable and what’s true. What mathematicians can prove depends on their starting assumptions, not on any fundamental ground truth from which all answers spring.
Since Gödel’s discovery, mathematicians have stumbled upon just the kinds of unanswerable questions his theorems foretold. For example, Gödel himself helped establish that the continuum hypothesis, which concerns the sizes of infinity, is undecidable, as is the halting problem, which asks whether a computer program fed with a random input will run forever or eventually halt. Undecidable questions have even arisen in physics, suggesting that incompleteness afflicts not just math, but—in some ill-understood way—reality.
Well, the 2022 Nobel Prize in Physics may just be another nail in the coffin for determinists.
The 2022 Nobel Prize in Physics was just awarded to scientists that just proved one of the more unsettling discoveries in the past half a century: the universe is not locally real. In this context, “real” means that objects have definite properties independent of observation—i.e. an apple can be red even when no one is looking. “Local” means that objects can be influenced only by their surroundings and that any influence cannot travel faster than light. This means that the influence of a particle can’t move faster than the speed of light. Investigations of quantum physics have found that these things cannot both be true. Instead the evidence shows that objects are not influenced solely by their surroundings, and they may also lack definite properties prior to measurement.
Many determinists held out the idea there were ‘hidden variables’ – or a lower level of reality we haven’t found yet – that would somehow be communicating between the particles and keeping the idea of realism (locally real) alive. However, in 1964 Bell released a paper showing that quantum mechanical behaviors do violate the idea that there could be ‘hidden variables’ – and even described the ways those violations would show up mathematically. What remained to be done was to develop an experiment to prove or disprove his assertions.
To disprove this idea of ‘hidden variables’ and prove Bell’s assumptions, they did this by using experiments on entangled particles that keep their state linked. Particles that are entangled (in this case photons with a certain polarization) and sent in two different directions yet still remain entangled in state.
They devised a clever experiment in the dungeons under Vienna’s Hofburg palace over the space of kilometers. They analyzed the results of passing these entangled photons through different filters and found that they do indeed adhere to Bell’s equations – and hence disprove particles adhere to the properties of being locally real (both at the same time).
The work does not prove which of those two principles (local or real) are false. Just that at least one (or both) is false.
Confused? Here’s a video that also describes the conundrum and what is going one really well:
Side note: Reading the state of one entangled photon determines the state of the other – but this state is always completely random (which is why faster than light quantum communication is not possible – you still need to compare the two results independently of the system to know if the random result was the ‘correct’ bit in the original message or the ‘wrong’ bit. Otherwise you just get a string of bits each random set to being correct or incorrect – which as it turns out is the only truly safe and unbreakable cipher).
Links:
Left bundle branch block is a problem with the heart’s electrical wiring (conduction) system.
Your heart has 4 chambers. The 2 upper chambers are called atria, and the 2 lower chambers are called ventricles. In a healthy heart, the signal to start your heartbeat begins in the upper right chamber of the heart (right atrium). From there, the signal activates the left atrium and travels to the lower chambers (right and left ventricles) of the heart. As the signal travels along the heart’s conduction system, it triggers nearby parts of the heart to contract in a coordinated manner.
Two bundle branches carry the electrical signal through the ventricles to the bottom of the heart and cause the ventricles to beat. These are termed the right bundle and left bundle. In left bundle branch block, there is a problem with the left branch of the electrical conduction system. The electrical signal can’t travel down this path the way it normally would. The signal still gets to the left ventricle, but it is slowed down. That’s because the signal has to spread from the right bundle branch through the heart muscle and slowly activate the left ventricle. So the left ventricle contracts a little later than it normally would. This can cause an uncoordinated contraction of the heart. As a result, the heart may eject blood less efficiently. For most people, this is not a big problem. But if you have underlying heart failure, left bundle branch block can make it worse.
Some people may have left bundle branch block for many years without any problems. But a newly diagnosed left bundle branch block may mean there is some underlying heart condition that requires prompt treatment. An aggressive evaluation may be necessary if you have new onset of a left bundle branch block.
Some people with left bundle branch block may need a permanent pacemaker. A pacemaker helps keep the heart beating at the correct rate. This is usually only needed if you are having symptoms or have another conduction problem along with left bundle branch block.
This was published in 2018, but it’s a fascinating dual purpose use of neural nets. Firstly, there was a massively increasing issue with character animation. Character animation is quickly becoming highly complex as it has becoming more realistic. The problem compounds when you want to make sure you can do things like crouch and aim at the same time. Or crouch and walk across uneven terrain while looking left or right. You can imagine all the different kinds of combinations of motion that must be described and handled. This all started taking massively more time to develop by artists; but even worse it was taking up more and more storage space on disk and especially in memory space.
Daniel Holden of Ubisoft wondered if he could use a neural net to not only reduce the combinations they had to handle into a net but also utilize the inherent nature of neural nets to compress data. It turns out he could – and he presents what he found in this excellent presentation.
Links:
Physicists recently use a neural net to compressed a daunting quantum problem that required 100,000 equations into a solution that requires as few as four equations—all without sacrificing accuracy.
The problem consists of how electrons behave as they move on a gridlike lattice. When two electrons occupy the same lattice site, they interact. This setup, known as the Hubbard model, is an idealization of several important classes of materials and enables scientists to learn how electron behavior gives rise to sought-after phases of matter, such as superconductivity, in which electrons flow through a material without resistance.
The Hubbard model is deceptively simple, however. For even a modest number of electrons the problem requires serious computing power. That’s because when electrons interact, their fates can become quantum mechanically entangled: Even once they’re far apart on different lattice sites, the two electrons can’t be treated individually, so physicists must deal with all the electrons at once rather than one at a time. With more electrons, more entanglements crop up, making the computational challenge exponentially harder.
One way of studying a quantum system is by using what’s called a renormalization group. That’s a mathematical apparatus physicists use to look at how the behavior of a system—such as the Hubbard model—changes when scientists modify properties such as temperature or look at the properties on different scales. Unfortunately, a renormalization group that keeps track of all possible couplings between electrons can contain tens of thousands, hundreds of thousands or even millions of individual equations that need to be solved. On top of that, the equations are tricky: Each represents a pair of electrons interacting.
Di Sante and his colleagues wondered if they could use a machine learning tool known as a neural network to make the renormalization group more manageable. The neural network is like a cross between a frantic switchboard operator and survival-of-the-fittest evolution. First, the machine learning program creates connections within the full-size renormalization group. The neural network then tweaks the strengths of those connections until it finds a small set of equations that generates the same solution as the original, jumbo-size renormalization group. The program’s output captured the Hubbard model’s physics even with just four equations.
“It’s essentially a machine that has the power to discover hidden patterns,” Di Sante says.
The work, published in the September 23 issue of Physical Review Letters, could revolutionize how quantum scientists investigate systems containing many interacting electrons. Moreover, if scalable to other problems, the approach could potentially aid in the design of materials with sought-after properties such as superconductivity or utility for clean energy generation.
Links:
The oft hinted Apple Car seems to have had some rocky development over the years with various starts/stops. The estimate is Apple’s next generation EV is projected to be revealed in 2024 with production starting 2025 – 2028.
At least some folks are coming up with designs for the automobile.
No field of science is littered with more mis-information and bad science than food and nutrition. Every year we hear about the latest fad diet – and every few years it turns out these diets are debunked. Even worse is foods that supposedly cure/fight condition <insert favorite disease here>.
Like many things, there is an element of truth – but the efficacy is in the details. Details that are often completely ignored. What’s a good example?
There are many top-selling anti-cancer books that preach the top food with cancer fighting properties to be broccoli and other cruciferous vegetables like Brussels sprouts, arugula, cabbage, kale, and cauliflower. But the devil is in the details; and the details are fascinating in this case.
In 2019, scientists were indeed studying the natural cancer-fighting properties of the PTEN gene that controls cell growth. People with good PTEN gene function tended to have better cancer fighting properties. During the research, they found that cancer produces the enzyme WWP1 that negates the natural function of PTEN and helps the cancer grow. While analyzing the WWP1 structure, they scanned databases of existing compounds that could bind and block WWP1. One popped up in the search, I3C. I3C neutralizes the WWP1 enzyme and lets PTEN operate normally in it’s cancer fighting role. It turns out that I3C is naturally found in, you guessed it, cruciferous vegetables like broccoli, Brussels sprouts, arugula, cabbage, kale, and cauliflower.
To test this theory, they injected mice who were engineered to develop prostate cancer with I3C. They found it greatly helped the lab mice. But there is a catch.
In order to reach the effective anti-cancer dose of I3C needed in a human, a person would need to eat 6 pounds of broccoli A DAY. This is why previous studies of cruciferous vegetables had good lab results but produced mixed results in humans. It implies the right way to go about this would be to develop a pill or supplement form for more easy consumption. Additionally, I3C affects more than just the WWP1, so additional work is needed to determine doses and unintended effects of I3C supplements.
So, always be sure to check on published, peer reviewed research before embarking on your miracle cure diet.
Links:
According to Tim Cook, Apple looks for these 4 skills/traits:
He seems to think of them in that order too. Collaboration is key because it encompasses the other 3. It is the fundamental notion that if I share my idea with you that the idea will grow and get bigger and better. That’s how Apple creates products. This creates a sense of teamwork which lends itself to creativity and curiosity – all things needed to improve or launch new products. We want people that are not caught up in the dogma of how something has always been solved. It’s amazing when somebody starts asking questions as a kid would.
I have found that questioning fundamental assumptions does sometimes lead to really unique re-thinking of problems.
Links:
