HNHacker News
TopNewBestAskShowJobs

robertskmiles

380 karma · joined February 28, 2011

submissionscomments
robertskmiles··on If You Have A Smartphone, Anyone Can Now Track Your Every Move
But when you're not using the internet, an idle Wi-Fi radio wastes power.
robertskmiles··on Outwitting the Prisoner's Dilemma
The biggest reason it's not a real prisoner's dilemma is because the prisoners can communicate. Key to the standard prisoner's dilemma is that there is no way for the prisoners to talk to one another.

Like the post says, it's a variant.

robertskmiles··on Introducing T: A command-line power tool for Twitter
The relationship terminology looks neat, but the ASCII table looks pretty messed up for me.

Maybe an HTML <table> tag is the right tool for the job here.

robertskmiles··on If You Have A Smartphone, Anyone Can Now Track Your Every Move
Likewise. In fact phones are more likely to encourage you to turn things off to save power.
robertskmiles··on Digital Design Principles from the UK Gov
> Unfortunately, the amount of difference this kind of thing will make is minimal.

Not true at all.

gov.uk is up and live right now in beta, and I have already used it to look up things about my taxes, some legal stuff for a friend, and visa-related things for a foreign acquaintance. It took no time at all; I put in the general search term, and the top hit is a simple, beautfiul page containing exactly the information I'm looking for, with links to all the resources I might need. Gov.uk is amazingly good. I am a guy who enjoys finding fault, especially in technological matters, but I would not change a thing about the current gov.uk site (except get rid of the beta warning).

So no, these principles are not binding for anyone else, but they're already making a difference. I hope that when people see how good government sites can be, they'll start demanding improvement. Either that or gov.uk will eat everyone else's lunch because people prefer to use it.

robertskmiles··on How to leave academia (for science PhDs)
> 100 passengers have queued up to board a plane, and are lined up in the order of the seats on the plane (n=1..100). However, the first person lost his ticket and selects a random seat. The remaining passengers will occupy their assigned seat if it is available, or a random seat otherwise. What is the probability that passenger 100 sits in seat 100?

It really worried me that I couldn't figure out how to work that out. I thought about what I'd do in an interview if I was asked that, and I figured (if I had my laptop) I'd write some code. So I gave myself 5 minutes by the clock, and wrote a little python program that simulated the situation and counted the results. I ran 5 million iterations, which with pypy took 59 seconds. The number I got out was a 97.335% chance of person 100 being in seat 100.

I have 3 questions:

1. Is that the right answer?

2. How are you supposed to work it out?

3. Would working it out the proper way take less than 6 minutes?

Edit: It seemed like too high a number, which is part of why I asked. I looked through my code and found a dumb error - I forgot to remove the seat from the 'available' list if the person finds their assigned seat. That's what happens when you write code on a 5 minute deadline. Now I'm consistently getting 49.9%, which seems more reasonable.

robertskmiles··on I'm Sure It Will Only Take You A Few Days To Code
Another thing which I think kills accuracy in estimating development time is the ratio of time needed for "figuring out what to do" and "actually doing it".

If you're learning a song, the first stage is very quick. You think about the problem for a moment and you decide on a plan like "Read the sheet music and try to play the song. Focus on areas you have trouble with. Repeat". The bulk of the time taken is spent "actually doing it", not "figuring out what to do".

The thing about the "figuring out what to do" stage is that often in order to estimate the time needed for that stage, you basically have to have already completed it. The answer could come to you out of the blue, it could take weeks of thinking into dead ends. And with software, that first stage can be the bulk of the time. Software is a more like writing a novel than learning a song. The thing that takes longest isn't the actual writing, it's the figuring out what to write. If someone asked you how long it would take you to write a novel that gives insight into the human condition by following the story of a family caught up in a bloody civil war, what time estimate would you give them? I expect your typing WPM wouldn't be a big part of the calculation.

You only know how long it's going to take once you know how you're going to do it, and once you know how you're going to do it, you've already done most of the hard work.

robertskmiles··on 10,000 People Sign Petition to Honor Alan Turing by Putting Him on the £10 Note
I like having Darwin on the £10, but I'd like to have Turing on a note as well. Maybe the £5 would be better, though that would mean getting rid of Elizabeth Fry, who I believe is the only woman currently on the currency (except Her Majesty of course), which may upset some people. The other options are the 20 and 50, which have Adam Smith and Boulton and Watt, respectively.
robertskmiles··on It’s all just bits
Nothing is "mere." --Richard Feynman

It's equally all "just quanta".

robertskmiles··on The pull-to-refresh patent
It's not that government-enforced monopolies on copying or using certain pieces of information are never justified, just that they aren't property.

Your right to drive as indicated by a driving licence is not 'property', your planning permission for a building is not 'property', your FCC permission to use a particular part of the EM spectrum is not 'property'. And your right to make copies of a particular piece of information isn't 'property' either. These are all abstract legal rights/privileges.

If you drive without a licence, broadcast in a band reserved for someone else, or copy a copyrighted work, then you are 'driving without a licence', 'breaking FCC regulations', and 'infringing copyright', respectively. Calling any of those things stealing is inaccurate, which is the kind of mistake you end up making if you think of them as property.

robertskmiles··on How to explain your game to an asshole
Would you have clicked on a link titled "How to explain your game"? I probably wouldn't have.
robertskmiles··on "Lighten up"
This is true but pretty unhelpful. The problem is that involving official proceedings or the law almost always looks like overkill. In the aggregate, these events amount to a very serious problem, but individually each one doesn't feel like enough to be worth making it official. Nobody wants to be known as "the person who got someone fired for a comment about a dress", even if they were totally justified. Even normal, decent people are going to be wary of you if you end up with that reputation, deserved or undeserved.

That's the essence of the problem I think. It's a big problem, but it's hard to do anything about it without being seen to be overreacting.

robertskmiles··on Is Intelligence Self-Limiting?
> If pleasure is its sole motivation, it will prioritize it over survival.

Which is why I proposed a design with no pleasure construct.

> If the AI can't change its own motivation (its utility function) it's nothing more than a clever automaton.

The utility function is the way the mind decides which states of the world are desirable. You don't need to be able to change that to be intelligent. I'm unable to change myself so that I consider my family being murdered to be a good thing, but that doesn't make me 'just a clever automaton'.

robertskmiles··on Peter Norvig: Tweaking Bayes' Theorem
Given accurate probabilities as input, Bayes always gives the best result. That's a theorem. So the fact that this tweak improves the performance probably means that either the english model probabilities are systematically too low, or the translation probabilities are systematically too high. My guess is the latter.
robertskmiles··on IPad as hip flask
My calculator did ruin my ability to do mental arithmetic, because I had it in my pocket basically every weekday between the ages of 11 and 17. Damn school blazers and their capacious pockets.

My general ability to do maths (in terms of understanding and applying mathematical concepts) is good, I got a degree in computer science, but my ability to do maths in my head is really bad.

robertskmiles··on Is Intelligence Self-Limiting?
> it can also modify the utility function to return MAXINT every time

Which would be equivalent to taking the murder pill. If it's able to model its own behaviour and model the consequences of future courses of action (required for meaningful self-modification and meaningful planning respectively), it will see that such a modification results in poor ore collection, and not make the modification.

You're right about the time-envelope of the utility function being an issue. The AI needs to plan far enough ahead at all times to see all relevant consequences of its actions. I don't think that requires two separate utility functions though, a single long-term one should do the job.

Edit: Also, "being able to modify the utility function is a pre-requisite to be called intelligent."? [citation needed]

robertskmiles··on Is Intelligence Self-Limiting?
> It's hard to balance the ability to create new, more useful utility functions with prevention of creating a utility function at odds with what the original entity valued.

I think that for such an AI, the concept of a "more useful utility function" would be a nonsense. The AI's definition of 'useful' is the utility function. No other utility function can ever rate higher against the current utility function than the current utility function does.

robertskmiles··on Is Intelligence Self-Limiting?
I think you're arguing a far tougher point than you need to. I was talking about the theoretical upper limit to intelligence. We don't need to make the argument that our particular civilisation will actually achieve it (though I agree with your post on that by the way); It's far easier to argue that hyperintelligence is possible in principle, and that's all that's needed to refute the point.
robertskmiles··on Is Intelligence Self-Limiting?
It's onomatopoeic, and capitalised not to indicate an acronym, but for emphasis. I don't know who originated the term but it was popularised by Eliezer Yudkowsky and Robin Hanson. The idea is that once an AI is able to self-modify to become smarter and consequently better at self-improvement, that feedback loop creates an 'explosion' of intelligence, and the AI "goes FOOM!".

"FOOM!" here is usually accompanied by some form of hand gesture evocative of an explosion.

http://wiki.lesswrong.com/wiki/FOOM

robertskmiles··on Is Intelligence Self-Limiting?
> Why would it want or not want anything, if it doesn't have a pleasure construct

Why does a calculator want to add, without a pleasure construct? If it doesn't enjoy giving the right answer to 2 + 3, what motivates it to choose 5? The answer is it doesn't necessarily 'want' to do it, it just does it. That was how it was programmed.

You program the AI to choose the plan that maximises the expected utility function. Someone looking at that AI in action might suppose it 'wants' to maximise the utility function. Whether they'd be right to suppose that is a question for the philosophers, but the point is, the thing doesn't just sit there apathetically just because it has no 'motivation construct'. You could say that the AI follows its programming, but it would be more accurate to say the AI is its programming.

robertskmiles··on Is Intelligence Self-Limiting?
It's not a straw man, it's a thought experiment. Obviously you can't actually do it. I'm sure you'd have trouble filling a room with rules about how to handle chinese symbols, or teaching a colourblind scientist everything it's possible to know about colour. It's hypothetical, it's a thought experiment.

Your first criticism is just invalid. The pill rewrites the workings of your brain to enjoy murder, the experience of murder and the consequences of murder. The fact that the brain originally evolved to work one way is irrelevant. The hypothetical pill changes the brain in a way just as powerful as the way an AI can rewrite its source code.

Your second criticism doesn't invalidate the experiment unless you claim that if the real life repercussions weren't a factor then you'd happily take the pill. I assume this is not the case.

Given all the facts, you won't choose to change what you fundamentally value, because that change would necessarily go against what you fundamentally value.

robertskmiles··on Is Intelligence Self-Limiting?
What's the purpose of this 'pleasure' construct in the AI's mind? If it's able to calculate the value of a utility function in order to set the 'pleasure' variable, and it bases its actions on the value of this 'pleasure' variable, why not just cut out the middle man and have it base its actions directly on the result of the utility function? The variable functionally buys you nothing, but introduces this problem that adjusting the variable directly can cause the AI to take actions inconsistent with its utility function.

Without the variable, the problem doesn't happen. The AI values collecting ore. If it has enough self-awareness to reliably modify itself, it knows that if it modifies its utility function it is liable to collect less ore, which is something it doesn't want. The action of modifying the utility function naturally rates very low on the utility function itself.

You don't want to murder people, so not only do you choose not to murder people, but if you are presented with a pill which will make you think it's good to murder people and take great joy in it, you will choose not to take that pill. No matter how enjoyable and good murder may be for you if you take the pill, your own self-knowledge and current utility function prohibit taking it.

The model of intelligence described can be thought of as self-limiting. Luckily it is not by any means the only viable model of intelligence.

robertskmiles··on Is Intelligence Self-Limiting?
Obviously intelligence tops out at the high end of a sigmoid curve - there's only so much matter and energy in the universe to convert to computronium - but that's no reason to suspect any value smaller than that as the limit. It must eventually level off, but the point at which it levels off may be hundreds of billions of times more intelligent than a human.

So the fact that intelligence is limited, doesn't in any way mean that hyperintelligence isn't possible. There is a limit, but we have no reason to believe that limit is on anything like the same order of magnitude as current intelligence.

robertskmiles··on Wicked Problems - Problems Worth Solving
I really like the idea of applying things learnt in entrepreneurship to dealing with social problems.

I often think that many charities and other altruistic organisations are stuck in various anti-patterns, some left over from as long ago as the Victorian era.

We've learnt a lot about getting big things done in the modern world; there's no reason that that expertise should only be applied to profit-seeking.

robertskmiles··on How To Build a Naive Bayes Classifier
The algorithm needs training data. In this example you have a load of emails where you've explicitly tagged each one as either spam or ham.

So you could perhaps do what you're suggesting, if you have a big database of emails tagged as 'interesting' or 'not interesting'.

robertskmiles··on How To Build a Naive Bayes Classifier
Yeah, probably. That's something that's fairly easy to measure for any given data set, so you can check to see if it's worthwhile before you do it.
robertskmiles··on How To Build a Naive Bayes Classifier
I found that note very interesting:

> consider "house" and "housing" ... the former is used less frequently in a spammy context then the later.

The point is that stemming is valuable unless P(spam|root) is substantially different from P(spam|derivedWord). But that data is available to you through Bayes. Has anyone tried a 'conditional stemming' system where the word is stemmed iff

|P(spam|root) - P(spam|derivedWord)| < someThreshold

? I wonder if that would improve classification accuracy. I suppose it might be a big performance hit on the stemming algorithm though.

robertskmiles··on Journals Warned to Keep a Tight Lid on Diesel Exposure Data
I expect the bigger HN-like sites could drive a substantial streisand effect here. I'm thinking of reddit in particular.
robertskmiles··on "DevOps is ruining my craft"
> I too consider myself an artist and a craftsman of server building. With each click of a mouse, I create a work of art.

I... You're setting up servers with a mouse?

robertskmiles··on Journals Warned to Keep a Tight Lid on Diesel Exposure Data
Streisand Effect though. Now people know that attempts were made to suppress the information, they are likely to take a much greater interest than they otherwise would, and the information will get a lot more attention and potentially do a lot more damage.
← PreviousPage 3 of 5Next →