mjw

18 ms·

121.
by mjw 14y ago
I don't think the point is to offer this literally as a likely outcome. If I interpret the parent correctly, the point is more to argue that there has to exist some level of inequality which even the most hardened libertarian would agree
122.
by mjw 14y ago
If you take a step back, one man's convenient is another man's inconvenient. And don't think for a minute that science limits itself to pissing off rightwingers. Nature has a way of pissing all over all sorts of ideological preferences. The
123.
by mjw 14y ago
I don't think it does support blanket tarring if interpreted reasonably. I don't see it as a positive thing when science gets suppressed because it's conclusions are considered politically incorrect or unfortunate -- either by the left or t
124.
by mjw 14y ago
Shame about the emphasis on long hours here. Not all start-ups expect them as a matter of course and I'm not convinced that they're a healthy part of start-up culture or crucial to success. People should be prepared to be flexible of course
125.
by mjw 14y ago
As an engineer who's investing in developing "deep expertise in statistics and machine learning" I can only stand to benefit from it, but something about the current wave of Big Data hype makes me instinctively a bit wary. Does this skills
126.
by mjw 15y ago
Yeah trigonometry does become pretty trivial if you're prepared to use complex numbers and some calculus. Here in the UK we were introduced to this approach later on at A-level, but still had to tackle an incredible number of "solve for the
127.
by mjw 15y ago
Surely a simpler explanation for that is that business people don't usually work with crisp, binary data. Instead they're (implicitly) doing some kind of statistical inference on noisy data. Propositional logic might be an acceptable approx
128.
by mjw 15y ago
This kind of thing is awesome in a way. I get the sense that machine learning really feeds on people attacking problems from both ends, the elegant probabilistic side and the practical optimisation hacks both inspire each other. Some potent
129.
by mjw 15y ago
No doubt there are some differences with the US here; the start-up career path perhaps isn't necessarily so widely known about and recognised outside the tech community, nor is it quite as breathlessly cheerled as in the bay area. But still
130.
by mjw 15y ago
I'm hoping the http://graphlab.org/ approach may help bridge the gap a bit here...
131.
by mjw 15y ago
I sadly missed a talk by its creator last week :( It looks interesting as a framework for parallelised algorithms where the dataflow isn't quite as straightforward as MapReduce (quite common in machine learning). I may try and use it for my
132.
by mjw 15y ago
OK, perhaps I should have qualified that. It's a good 90% solution for a lot of use cases, although obviously more useful for implementing algorithms in terms of numerical building blocks, than ultra-performance-critical code where you want
133.
by mjw 15y ago
I wonder what they think about or have learned from http://en.wikipedia.org/wiki/Fortress_(programming_language) , another recent-ish attempt to deliver a modern and powerful scientific programming language. Personally I'm a little wary o
134.
by mjw 15y ago
Having taken some courses from David and others at UCL recently, I'm a big fan of this. The Bayesian modelling perspective I think is very useful if you're interested in machine learning as more than just a collection of clever algorithms a
135.
by mjw 15y ago
Great observation; as an undergrad, noticing this suddenly made GIT (which was taught to me by a mathematician in all its intricate detail) seem a lot less scary.
136.
by mjw 15y ago
Definitely this. It may be counter-productive to over-analyse these things too, at least beyond a point. I ran myself into a bit of a dead end with jazz piano trying to be ever more inventive when it comes to harmony. I probably thought abo
137.
by mjw 15y ago
Depends how far along you are and what sort of programming you want to do really. Discrete maths and logic are probably the main areas of maths useful for computer science. Maybe look for a book on discrete maths for computer science. Set t
138.
by mjw 15y ago
I think as the employee gets a larger and larger influence over decisions about their own compensation, the line between 'failure to do your job correctly in the best interests of your employer' and 'embezzling funds' becomes more and more
139.
by mjw 15y ago
As someone doing a second master's degree in a mathematical field, I can't say I've found it 'mostly easy', or even mostly approachable. It can be fun once you've developed strong intuitions for the material, but often a hard slog to get th
140.
by mjw 15y ago
(To clarify, I did have fun there and learn quite a bit; if you like the work, you've got to take that into account too when considering opportunity costs vs working at a bank etc. But my advice on options still stands.)
141.
by mjw 15y ago
Once: nothing. (Except a mac mini which the founder was nice enough to buy me as a kind of `exit bonus' :) Despite my options having vested, investors held preferred stock which (since the exit was small-ish) left no liquidity on the table
142.
by mjw 15y ago
For the programmers out there, it's a bit like the difference between writing code which takes concrete data types ('n-tuple of floats', 'n by m matrix of floats'), and writing more generic code which takes abstract datatypes (roughly: 'thi
143.
by mjw 15y ago
> I should also point out that mathematicians don't number at zero unless there is some advantage (the default is to start at one) I disagree. While this may depend a little on which areas of mathematics you study, I've found that in si
144.
by mjw 15y ago
One advantage I think the parent is alluding to, is that zero-based indexing makes it a bit easier and more natural to generalise between continuous (functions on R) and discontinuous (functions on N) situations. Since the latter are often
145.
by mjw 15y ago
I'm inclined to agree. I've tried a few times to learn category theory in the abstract, and struggled to motivate myself when I couldn't clearly "see" the structure and the application, in categories (like Hask or Vect) which I know reasona
146.
by mjw 15y ago
Are they though? in which language? I was under the impression that the traditional 'catch' block has the stack unwound for it back to a frame with the same lexical scope as the associated 'try'...
147.
by mjw 15y ago
Fair enough -- I forgot about 'self' being dynamically scoped, which is an interesting point. Perhaps it's no coincidence that I have a particular dislike for the scoping rules around Javascript's "this" keyword, as they're perhaps even mor
148.
by mjw 15y ago
Unless I'm getting the wrong end of the stick, this appears to use dynamic, rather than lexical scope for the special non-stack-unwinding exception handler blocks. To me non-lexical scope is quite confusing, especially if you're proposing a
149.
by mjw 15y ago
Glad I'm not the only one bugged by this. Next time I think I might try the http://rubyforge.org/projects/rsruby/ bindings instead.
150.
by mjw 15y ago
I'd love to see an "R for rubyists/pythonists/...", basically an R tutorial aimed at reasonably experienced programmers which explains what you need to know about the quirks of R to properly grok it as a language, how it differs to ones you

More ›