Posts

Showing posts with the label artificial

I, Robot

"Gloria had a grip about the robot’s neck that would have asphyxiated any creature but one of metal and was prattling nonsense in a half-hysterical frenzy. Robbie’s chrome-steel arms (capable of bending a bar of steel two inches in diameter into a pretzel) wound about the little girl gently and lovingly, and his eyes glowed a deep, deep red." - If you didn't recognize the narrative, it is from Gloria & Robbie's reunion from the touching ending of Isaac's "I, Robot" first story. If you read "Robbie" before, you are probably, by now, recollecting what actually preceded this very moment of two persons getting together in this happy ending of the most famous Asimov short story. But if you never did, I am encouraging you to do so; if nothing else, then for the simple reason that even though it was written some eighty years ago, the premise is still fresh and valid, just like it was published yesterday.


The word 'robot' was actually coined a couple of decades before 'Robbie' by the Čapek brothers, Karel and Josef, and was first used in Karel's play 'Rossumovi Univerzální Roboti' (Rossum's Universal Robots). Although robots in this play were more androids and cyborgs, in fictitious literature they were not fully mechanical, AI-type automated inventions but rather sort of technologically augmented humans or created biological organisms. Nevertheless, the word stands to this day with its root in all Slavic languages, Serbian included. I remember my grandmother often using the word 'rabota' ('robota' in Czech), which, even though not in use in the official Serbian language, is actually the only possible way to represent the hard work or labor in just one word.

In sci-fi literature and motion pictures, not all robots are equipped with artificial intelligence, emotion chips, and sophisticated technology and created to look exactly like we do. Many of them are made just to do hard work, like in the original Isaac's or Karel's stories, but even though they all have one thing in common. Their own personality. Something that makes the robot unique and has properties only living organisms have. Believe it or not, if you have a vivid imagination or perception of the details of your own surroundings, personality is something even ordinary items can own. Not long ago, when my son was at the appropriate three-year-old age, I bought for him a large helium balloon to play with indoors. At the end of the meter or so of string, I hung a couple of iron rings to weigh just enough for the balloon to float freely in the air. It was fun playing with it, of course, but even more fun was just monitoring what it did on its own. Due to invisible drafts and air circulation in our flat and slight differences in pressure and temperature in different rooms, it was obvious that our 'Balloon Boy', as we called it, never wanted to stay still for a long time, and after a while I noticed that it particularly liked the kitchen. No matter where you floated it initially—in the living room, dining room, or hallway—after a couple of hours it drifted away to its favorite spot and stayed there put. And to do so, it had to pass through several corners and doors and avoid solid items and furniture. Now you tell me, how was our Balloon Boy different from any other home pet? It had a name, it required constant attention (instead of feeding, in this case, adjusting weights to compensate for helium lost), it also had its own favorite spot in the flat, it loved to play and drift, it was cool, and it... well, eventually died. From my point of view, Balloon Boy was no different than any living pet, and with all of his regular activities, 'he' earned his own personality. Not a big one, for sure, but personality-wise it was.


Robbie was designed to serve as a nursemaid, but in the end, from one young girl's perspective, it was a perfect pet or Balloon Boy substitute. He didn't talk but was able to mimic all the personality necessary to be an ideal companion for eight-year-old Gloria. And he was a great listener, something parents nowadays rarely have time or patience to do for their children on a daily basis. Robbie was also the first robot in Isaac's "I, Robot" masterpiece and surely one of the first generations of robots. With later stories and the overall sci-fi genre, within robotics and cybernetics naturally comes artificial intelligence. In this realm, my favorite robot in the entire expanse of science fiction is Commander Data from "Star Trek: The Next Generation". He has it all and was as fully functionally self-aware as anyone else in the franchise. But we are far away from such achievement. I mean, creating artificial software design to mimic human beings within the current stage of hardware and software is very much possible. Computers are fast enough to process very detailed responses from the surrounding environment. Sensing tools are also mature enough to visually and audibly acquire all the data for a hypothetical humanoid robot to deal with and to be very close to passing the Turing test. Simply put, I am convinced that very soon we will have artificial Facebook contacts you can add to your friend lists and communicate with in the usual manner and never know that they are not really humans. To be perfectly honest, I will not be surprised if they already exist today and use social networks as a perfect beta testing ground.

However, what is still behind rapid development in computer science is power and mechanics. These days Boston Dynamics' Atlas's new upgrade is going viral, and if you haven't already seen what it can do, please take a couple of minutes to watch the above video. It is amazing what they achieved in only a couple of years of development from the first 'Petman' bipedal robot initially constructed for testing chemical protection suits. Still, even though walking and handling simple labor is vastly improved, motion and sophistication are yet to explode in some sort of technical breakthrough that would allow continuous operation without the need to recharge often and, of course, to have more human-like motion abilities and be able to do various actions, from as sensitive as operating smartphones to as bulky as carrying heavy sacks and boxes. And at the same time, to look like one of the chess players from the above photo. Or both of them. Until then, there will be no fear of some futuristic robot uprising in Boston Dynamics, especially against those test people from the Atlas video.


To conclude this post without mentioning industrial robots would not be really fair. They have been among us for years and doing their job with great perfection. Honestly, one of the 'always on' TV channels playing in our living room is 'Discovery Science', and I simply can't get enough of those shows "How It's Made" and "How Do They Do It?" especially with all those automated industrial lines with heavy usage of robots and machines. Cybernetics is one great engineering, and it literally expanded exponentially with microcontrollers and industrial software. With a little regret of missing the opportunity to pursue a career in robotics, I remember those couple occasions in school when I participated in a competition in building a controllable circuit board with the Sinclair ZX Spectrum, an amazing one of the first home computers I owned for years back in the eighties. It had an extension connector, designed for accessories, with 8 pins fully controllable with its famous PEEK & POKE commands. I used it to programmatically control electronic relays capable of controlling the flow of heavy electronic AC current.

Now, even though industrial robots are already ready to go to another planet (and a couple of them have already been sent to Mars and successfully did or still do their jobs), androids are still not on the horizon. Should I dare to predict the first commercial humanoid robot on the market? Let me put it this way: the human body contains 200+ bones, 600+ skeletal muscles, and more than 300 joints. When we reach a scientific breakthrough in using artificial muscles and power systems able to operate a vast number of joint and bone movements, it will not be long before we see the first Robbies in hardware stores.

Image credits:
http://www.templates.com/blog/robots-people-striking-3d-perspective/
http://www.blutsbrueder-design.com/
http://dailyinbox.com/next-decades-manufacturing/

Is Life a Zero-Player Game?

Think about it. If life really is some sort of game and we are just characters in one giant artificial intelligence play, then... Well, let's just say that we can safely recognize not very enjoyable rules we unconscionably must obey. They are simple. We must play the game. We can't quit the game. We can't win. Oh, and yes, if life really is a game, then we are only either slaves in one master-puppeteer god-like performance, or we could be just a bunch of units interacting with each other in a sort of limited free will world or a world where free will is just an illusion. Now, if life really WAS a game, what would you prefer?

Olivia Wilde & Garrett Hedlund in 'Tron: Legacy'*

It is obvious that the first option is what we easily recognize as a religious world. If you ask me, this is a simple marionette type of world in which we, being game units, have little or no influence in the game, and we must obey divine rules and please the puppeteer. From my point of view, let's just hope this is not the case. However, the second scenario is something worthwhile to give further thought to. If life is something like one large simulation with characters playing the game independently without creator influence during the game, then we are just participating in one giant zero-player environment that started eons ago in the point of history where evolution began with a predefined start pattern. And evolution is nothing more than just a set of rules in the complex game algorithm, and time is just an iteration flow in patterns changing from one state into another by following the rules.

Confused?

Ok, let's simplify the scope and check one famous zero-player game that might help understanding the basic principle. The inventor is perhaps one of the great minds in the world, John Horton Conway, a mathematician from Princeton University, who tried to simplify the original John von Neumann idea to explain evolution with the creation of a mathematical model without explosive growth over time, using just small initial patterns with unstoppable and unpredictable outcomes with a set of rules as simple as possible, which would drive the entire system forward in time. Conway came up with a brilliant two-dimensional matrix where one dot represents one living cell. Cells obey four simple rules:

1. Any live cell with fewer than two live neighbors dies (death by underpopulation).
2. Any live cell with more than three live neighbors dies (death by overpopulation).
3. Any live cell with two or three live neighbors continues to the next generation (survival).
4. Any dead cells of exactly three live neighbors will come back to life (birth).

Conway's matrix is, just like life, infinite in size, but for demonstration purposes the following example is finite matrix that. Just FYI, as a single cell is clickable, I couldn't make it mobile or any small display friendly. To try it out, you would really need to use an old-fashioned computer screen and mouse pointer. Please feel free to play and create your own pattern and see what happens. This is, of course, a zero-player game, so your godlike intervention in this game is only to create the initial organism. The Conway's game of life then operates on its own, and you can only watch.


Conway's brilliant experiment is only a two-dimensional game with a small number of simple rules, yet it opens endless fun and endless variations in the evolution of different patterns and their interactions. Now, is it possible to create, hypothetically speaking, a very complicated game on a molecular level with complicated rules within the realm of chemistry? And instead of an endless matrix, use the three-dimensional surface of a planet? Is that what the Earth is? One giant playground with molecules in endless interactions with each other, and we are today just a snapshot in the game's current evolution stage?

It surely fits the world surrounding us and the one in the past. In this game, the world before was less complex than it is today, and the world today is less complex than the one from the future. Living units in the game are evolving due to infinite interactions, and if we go to the very beginning, to the first pattern of living cells, some 3.8 billion years ago, approximately 750 million years after Earth was formed, it is clear that we indeed might be living in a complex biological game. The game is without players and puppeteers and only with living organisms with developed conscious minds. In order to neatly describe the current stage of the game, I will just quote Stephen Hawking: "We humans are highly complex biological machines behaving in accordance with the laws of nature. Our brains create and sustain our conscious minds through an extraordinary network of interacting neurons. That consciousness creates a three-dimensional model of the outside world: a best-fit model that we call reality."

Red Pill or Blue Pill?***

You might be asking now where free will fits in the game. If we are not players per se, then do we even possess such things? Are we able, being units in the game, just by following the rules, no matter how complex they are, to choose our own course of action without constraints and fate? If the game model like this one is correct, then I am pretty sure we can stop thinking about free will. There is no such thing, at least in the raw meaning of the word. Yes, we are able to control our actions and to choose certain paths, which gives us the illusion of free will, but even if we choose one path in favor of another, we are not really capable of calculating where this chosen path really leads to or where it ends. There are simply too many unknown variables on the way. Not to mention that we are completely incapable of knowing who or what we will stumble on on the chosen path and how this new interaction will play out in the game.

But the beautiful thing in this mind experiment called "Game of Life" is that even though we only have limited free will, as it seems, there is no fate as well. And even though the rules are definite and inexorable, due to the enormous size of the game level and complexity of the rules and the infinite number of organisms and molecules, it is really impossible to calculate the outcome of the game or any of the game's parts separated either in space or in time. At least from the inside of the game. And as it appears, there is no outside of the game as well. If there was, then, like in Conway's game embedded in this post above, there could be a "reset" button somewhere. "The button" that has perhaps been pressed about five times so far.****


But, like in any game, there might be glitches, lags, and bugs (like fabulously portrayed in Tron movies and series*). And I definitely had that in mind when last summer Viktor and I filmed a short movie with the same name** that exploits this very scientific thought. It's our first and only movie so far, so it's full of imperfection, but to sum it up, its plot tells a story about a young boy who's following a glitch in the system, presented in real life as a firefly, through numerous portals to the place where he meets a man with the final orb, the artifact that seems to be a way in for full understanding of life itself, its origin, and the rules it is built on. The entire movie is embedded above, and for more about all the filming and production, please find the referenced link within.

Image refs:
http://www.imdb.com/title/tt1104001/,
** http://www.milanzivic.com/2016/08/game-of-life.html
*** Cornell Math Explorers' Club

Code ref:
** https://codepen.io/RBSpatz/pen/rLyNLb

Refs:
**** http://www.milanzivic.com/2015/06/the-sixth-great-dying.html
http://www.bitstorm.org/gameoflife/lexicon/
https://en.wikipedia.org/wiki/Conway_Game_of_Life
https://en.wikipedia.org/wiki/Zero-player_game
https://en.wikipedia.org/wiki/John_Horton_Conway
https://en.wikipedia.org/wiki/John_von_Neumann
http://www.hawking.org.uk/videos.html
http://www.physicsoftheuniverse.com/topics_life.html
http://www.ncbi.nlm.nih.gov/books/NBK9841/
http://www.iep.utm.edu/freewill/