Posts

Showing posts with the label prediction

Brain #1: Purpose and Optimization

For what purpose has the brain evolved? As humans, we are biased when we encounter that question, writes Lisa Barrett in Seven and a Half Lessons About the Brain . We wrongly assume that the purpose of the brain is to think: “After all, thinking is the human superpower, right?”   Wrong, says Barrett. Long, long ago, unicellular life found itself in competition with others over limited resources, the importance of any capability to sense what lay where was an evolutionary advantage – Did XYZ lie to the left or right? Gradually though, raw sense organ signals weren’t enough. Choices had to be made – was it likely one could catch the prey? Make a wrong choice repeatedly and one would die of starvation. Thus: “Energy efficiency was a key to survival.”   So Barrett concludes: “Your brain’s most important job is to control your body… by predicting energy needs before they arise so you can efficiently make worthwhile movements and survive.”   But this created...

Give me a Date

I was talking to this Christian friend of mine and the conversation turned to Biblical prophecies. He talked about the impending Armageddon. As he told me jokingly, that’s not Armageddon as in the movie; rather it’s Armageddon as in “the last battle between good and evil before the Day of Judgement”. Ever the detail oriented one, I asked, “What’s the date for that?” As you might expect, he had no answer to that question. Instead, he said it would happen right when Israel was surrounded by everyone else in the world (not just the Islamic world). But that doesn’t give me a date either, does it? To me, this sounds like the “turtles all the way down” answer! This unwillingness to give a specific date is, of course, not just limited to religious folks. Seth Godin avoided a date when he talked about believing in yourself and moving forward with what you believe to be right: “Sooner or later, your work speaks for itself.” Worst of all, though, are the economists and financ...

Experts and Brexit

Was the Brexit vote the “wrong” choice? The experts certainly think so; but then again, did the experts see the financial crisis of 2008 coming? If they couldn’t see the greatest financial catastrophe since the Great Depression, should we really believe their doomsday predictions now? In Thinking, Fast and Slow , Daniel Kahneman said that the reason experts fail may not be arrogance. Rather: “The reason is that the person who acquires more knowledge develops an enhanced illusion of her skill and becomes unrealistically overconfident .” Overconfident and arrogance: it’s a critical difference. Arrogance is without basis; overconfidence is a problem of degree. Note how this rules out the seemingly obvious solution of calling oneself useless. If you consider yourself terrible, would you have any confidence in your forecasts?! In fact, as Philip Tetlock wrote in Superforecasting , what we need is a different kind of humility: “The humility required for good judgment is not s...

Predictive Search

I read this recent blog by Nicholas Carr (he of the “ Is Google Making Us Stupid?” fame) criticizing Google for its Autocomplete feature (you know where you start typing and Google shows you likely ways you are likely to finish that query) and now its attempt to give you an answer before you even asked the question! So what is Carr’s objection to Google’s prediction algorithms? His minor grouse is that: “Google’s goal is no longer to read the web. It’s to read us.” I am guessing Carr doesn’t like his mind to be read. Even if it’s only by an algorithm. As an aside, if we are that predictable, doesn’t it open up questions about free will? But ok, that’s a topic for another day blog… Moving on to his major grouse: he quotes Eric Schmidt, who in 2006 when he was Google’s CEO, felt that Google should “tell me what I should be typing.” Imagine that: to give you an answer before you even asked the question! How awesome would that be? Well ok, that feels a bit creepy too. ...

The French Can Predict the Future

Heard of this ruling by a French court asking Google to pay damages and a fine to a French map making company? “A French court has ruled that Google's free Google Maps application is anti-competitive and has ordered the company to pay €500,000 to Bottin Cartographes, a for-pay map company, as well as a €15,000 fine. Bottin Cartographes argued that Google was only planning to give away the service for free until all the competitors had been driven out of business and then they would start charging.” I don’t even know where to begin. So had only Google charged for its maps, there would have been no problem! Second, I thought justice is about punishing people who committed a crime, not punishing based on opinions about what they might do in future (Like Google first driving all mapmakers bankrupt by giving maps for free and then starting to charge for them). That sounds like a prediction to me. Have the French got some Minority Report style technology to predict crimes of the fut...