From MSc to PhD in 2 year's time... in Spain this is the time you are expected to figure out what to research about. No doubt why Spain needs a bailout. So much to learn.
A very practical and useful collection of hints. I particularly support the point about tailoring the set of informative features. In the end, there is no free lunch for the "core" of the machine learning method (cf., http://ti.arc.nasa.gov/m/profile/dhw/papers/78.pdf).
Now deep learning seems to debunk the more-or-less usefulness of the feature extraction step.
And the closest to solving this we seem to be at present is fuelled by deep learning, which is basically a big neural network with an absurdly vast amount of neurons (i.e., parameters for learning, like the brain). We can observe how this brute-force technique works, but unfortunately no-one can explain why (it's a black box model). The same story has been on for decades.
I would relate it to being a discriminative model, which is tailored to solving a specific task, in contract to generative models, which try to model and explain the world. Perhaps the brain is not meant to understand how the world works but how to do take advantage of it.
Deep learning models definitely don't need an "absurdly vast amount of neurons" - for example, GoogLeNet (arguably the state of the art image classification model) has only ~6M parameters.
I would still add another challenge for analogue: power harvesting. That's an environment where the amount of energy you can collect is so low that you can't simply afford to run a uC. However, you still need to manage that negligible amount of energy, because in the long run, it builds up, and then awesome stuff happens (switching a uC on, for example).
I bought the 2nd edition 15 years ago, I loved it, and I now cherish that useful book in my bookcase.
A wonderful book, indeed. It "teaches" a useful business management lesson: when two strong egos clash, the company drowns.
I also admire Carmack's way to tackle new (sometimes unknown) problems: read the literature, learn, do. And never failed following this.
I loved the parts when Romero swam across the lake to work all night with the rest of the crew, or when they invited in a stripper with pizza but Carmack wouldn't set the keyboard aside. So determined.
>I loved the parts when Romero swam across the lake to work all night with the rest of the crew,
That was also one of my favorite passages:
"The lake house was filled with the sense of unlimited possibilities. And
the bond between Carmack and Romero was becoming stronger by the day.
It was like two tennis players who, alter years of destroying their competition,
finally had a chance to play equals. Romero pushed Carmack to be a better
programmer. Carmack pushed Romero to be a better designer. What they
shared equally was their passion.
This was most clear to Carmack one late weekend night. He was sitting in
the house working at his PC as lightning flashed outside. Mitzi curled lazily
on top of his monitor, her legs draping over the screen. The heat of her body
was causing Carmack’s heat-sensitive display to ooze its colors. He pushed
Mitzi gently from the monitor, and she scurried away with a hiss.
A rainstorm had picked up, and it was mighty. Cross Lake spilled into the
backyard like the prelude to a horror movie. The lake was so high that it
pushed the ski boat to the top of the boathouse. Long black water moccasins
slithered toward the deck. The bridge leading to Lakeshore Drive was completely
washed out. When Jay arrived after having been out for the day, there
was no way to get in. It was, as he described it, “a turd floater” of a storm,
bringing everything from the bottom of the lake to the surface. He turned
away to wait it out.
Romero arrived with a friend later to find the bridge even worse than
when Jay got there. There was simply no way he was going to get his car over
the flooded expanse. And there were probably alligators and moccasins now
making it their home.
Back in the house, Carmack resigned himself to working on his own that
night. After all these hours, he had come to appreciate Romero’s diverse range
of talents, gleaned from years of making his own Apple II games. Romero
had been not only a coder but an artist, a designer, and a businessman. On
top of all that, he was fun. Romero didn’t just love games; in a sense, he was a
game, a walking, talking, beeping, twitching human video game who never
seemed to let anything get him down. Like a game character, he could always
find an extra life.
Just then the door behind Carmack swung open. Mitzi dashed under his
feet. Carmack turned to see Romero standing there with his big thick glasses,
soaking wet up to his chest, lightning flashing behind him, a big smile on his
face. It was a real moment, a moment so impressive that Carmack actually
saved it in his thin file of sentimental memories. This one he wanted for
future access: the night Romero waded through a stormy river to work."
Precisely. In fact, I chose the term "maker" for my blog: http://ai-maker.com/ and the gist of it is teaching, discussing, critiquing and creating stuff related to Artificial Intelligence (in this particular case). To me, none of the aforementioned topics is incompatible.
Off topic: Self promotion is not prohibited on HN, in fact it's even (semi) encouraged. But I feel like I've seen link to your blog too many times lately. And indeed, looking at your comment history, it seems to me that every other comment is a content-free plug to the blog. That's a bit too excessive
I like seeing (soft) AI as the buttress that allows us to see farther, like Newton standing on the shoulders of giants. I guess the sudden presence of AI is due to the sudden amount of available data (and thus, a potential source of useful information).
It must be hard to resist a big payslip, we're all humans, heads of family, we want to provide the best for our beloved ones.
I see a veneer of disdain to Norvig's attitude in your words. I don't know much about this criticism, but to me he has done a lot of good to us researchers with his AIMA book. In fact, this is the core that vertebrates my blog: http://ai-maker.com/. In the end, Goertzel (guru in AGI) works again for a financial prediction firm, doesn't he?
