It's an inescapable reality that many of us humans are embedded in a world where we aren't one of the physically beautiful people. I suspect it's more difficult for people who are borderline beautiful. Fortunately, I have less of an issue! I accepted my standing early in life.
In social media, we are constantly bombarded with images of physically beautiful people. The kind of people that we don't often see in real life (unless you live in a big city). So what's one to do if she's constantly reminded of her lack of perfection?
We are evolutionarily wired to notice beauty (btw, what's considered beauty is in the eye of the beholder). It's an unconscious thing that constantly creeps up into one's consciousness. This attention device is omnipresent.
The wonderful thing about education is that you can discover beauty elsewhere. The stuff that occupies my mind rarely involves human physical beauty. I'm fascinated by abstract beauty. But the truth is, it isn't as primitive an instinct as human beauty.
It's not just human beauty that we are attentive to. Those other cute and cuddly mammals are also the kinds of images that keep popping up on social media. But at least in our everyday existence, these pets are more accessible than the humankind.
It's fascinating that birds have advanced to a state where they've "invented" all kinds of different beauty norms. Birds are unencumbered by two dimensions. There is a likely causal relationship between mobility and the diversity of beauty norms.
Perhaps there is a fundamental theory about beauty norms. It's different from the notion of beauty is in the eye of the beholder which is also incontrovertibly true. That there are also cultural norms of beauty that are established, but mobility generates greater diversity.
So it should never come as a surprise that as our human civilization becomes more complex and affords much greater freedom and mobility of its citizenry that the norms of beauty become more diverse.
The side-effect of what is found beautiful is the attachment of value. It's different from the value that is attached to things of utility. In a modern civilization what becomes valuable are things that can lack utility. It should be no surprise then to discover expensive art.
Birds go beyond the selection of physical beauty and often select based on the exhibition of complex cognitive behavior like singing. Similarly, humans are also attracted to people who can make us laugh (and hence bring us joy).
If we say that beauty is in the eye of the beholder, then we must acknowledge that how we think of the world influences how we see beauty in the world. The more complex our thought processes, the more abstract our sense of beauty.
Thus to see all the beauty that reality has to offer, one has to train and construct our cognitive scaffolding such that we can recognize and then create beauty. The ultimate goal of humanity is to encourage this ability for everyone.
What is good is for everyone to know how to see beauty in this world. What is evil is to trap people into seeing everything as ugly.
There's a mad rush in Machine Learning circles to say 'X is enough' or 'X is all you need'. We have 'Rewards is enough', 'Attention is all you need', 'Diverse training is all you need', 'Size is all you need' (Bitter truth/Scaling hypothesis).
These logical jumps are ignoring all the messy details. I'm guilty myself of some ideas like 'Intuition machines is all you need' or 'Empathy is the path to general intelligence'. These singular ideas are pleasurable because they make what's complex appear simple.
Brains and Deep Learning are both live-wired systems and they are anything but simple. They are complex adaptive systems and require the entire kitchen sink of technology and mathematical models to get a handle on.
Millennials are pretty much f**ked. 911 created a jobs program for the military industrial complex, sending investments outside of the US. The real-estate crash led to government austerity and hence less opportunities. The pandemic was the nail in the coffin.
Now it's their responsibility to clean up the environmental mess that previous generations left them. They have to do this while no money is spent on there because previous generations want to retire comfortably.
But what are Millenials doing? They are focusing on cultivating their brand on social media. What is often the case is they avoid conflict and keep silent about the injustices that they are subjected to.
The way modern civilization is rigged, it is no surprise that the greatest minds of our generation are wasting their lives hacking the system for profitable vulnerabilities.
There is a difference though with hacking nature and hacking human nature. It is just sad to see that most of society spends most of their lives hacking the latter.
This is because modern civilization involves humans in games of one-upmanship. Humans are rewarded in proportion to the number of people they can influence. We are rewarded by the quantity of influence and not the quality of influence.
In Apple's rendition of Asimov's Foundation, the empire is ruled by 3 clones of the same original ruler. These clones are at different ages where the middle-aged clone is the ruler, the elder is an advisor and the younger is his successor.
In addition, there is a character Demerzel that serves the trio of clones. She is ever-present with the clones and all their ancestors. From singing lullabies to them before their birth to sending them off to incineration in their death.
Demerzel is immortal because she's an android. In Asimov's Foundation universe there are no AIs with this exception. Apparently, an AI is always present that is serving (or perhaps manipulating) the rulers of the civilization.
One of the most dangerous afflictions of data science teams is to go really big for the sake of going really big. That is why everyone jumped on the Big Data bandwagon and got little ROI to show for it.
Yes, Microsoft and Nvidia have the compute resources to go very big (i.e. 530b parameters), but that doesn't justify that everyone else does the same thing! microsoft.com/en-us/research…
What do you call that cognitive bias where you believe that you cannot make good progress without the fastest most advanced piece of hardware? This affliction affects so many technical endeavors. We all want to play with the F1 cars that everyone raves about.
The academic community would like one to believe that a single AI training method can lead to a useful system. This belief is not even remotely true. Indeed it gets you to publish a paper, but a useful product is very different from an academic paper.
A useful product is one that can be operated economically and addresses a user's needs at the correct price point. There are a multitude of knobs to tune here and a multitude of methods with varying resource demands, latency and accuracy.
A one-size-fits-all solution is a fantasy when it comes to products driven by AI methods. To deploy the right product requires a balance of many existing methods. This kind of balancing act is extremely difficult to do if we have tunnel vision of what methods are available.