HNHacker News
TopNewBestAskShowJobs

hgibbs

186 karma · joined October 6, 2013

submissionscomments
hgibbs··on Show HN: Floptimal – Preflop ranges reimagined
Implicit in a range like this one is that you are facing 'maximally exploitative opponents'. Since the game is zero-sum, over enough games the expected value of such a strategy is 0. However, since that expected value is against a maximally exploitative opponent any deviations of the opponent's strategy lead to an increase in your expected value. This is the game theory approach (GTO).

On the other hand, you can try to tailor your strategy to be maximally exploitative of the other players at the table, but doing this can be hard since often it can take a large number of hands to gauge what kind of player somebody is. On the other hand, GTO strategies are naive in the sense that you don't need to know anything about your opponents to guarantee that you Breakeven.

hgibbs··on Stability of Fixed Points of High Dimensional Dynamical Systems
The study of dynamical systems from a probabilistic perspective is called ergodic theory. Have a look at my other comment for some specific references for more on this.

And yes it is very interesting. One can actually model iid coin flips (or dice rolls) as a type of expanding dynamical system called a full-shift.

hgibbs··on Stability of Fixed Points of High Dimensional Dynamical Systems
The most accessible introduction (in my opinion) is 'Laws of Chaos - Invariant Measures and Dynamical Systems in One Dimension' by Boyarsky and Gora. For a short survey on just stability of the invariant measure you could look at 'Linear Response, Or Else' by Baladi. I also like the lecture notes 'Statistical properties of uniformly hyperbolic maps and transfer operators' spectrum' by Carlangelo Liverani.
hgibbs··on Stability of Fixed Points of High Dimensional Dynamical Systems
Some of the conclusions in the post are misleading. For systems that are uniformly expanding the statistical description of the dynamical system is very stable, and such systems can have sense periodic orbits that are necessarily all unstable! By statistical description I mean (at least) the long term frequency that the system is in any particular region of state space (i.e. the systems physical invariant measure).

It is actually one of the nice facts of life that as you dial up the local instability of a dynamical system you get a corresponding increase in the statistical stability. As systems become more and more chaotic they act more and more like iid coin tosses, which we understand quite well.

hgibbs··on Strategies to Reduce Brain Fog and Improve Clear Thinking
"Check that you aren't dehydrated" ought to be on any such list.
hgibbs··on Jim Simons proved the textbooks wrong, almost
Reversal/conversion on long dated options.

Not a pure ARB bc of dividend and interest rate risk, but it's close.

hgibbs··on What Is a Gamma Squeeze in the Context of Stock Trading?
The reason not to trade something that cannot be hedged is because it is impossible to replicate. Of course if the future cash flows of the security are known in advance this point is moot, but for options the idea is that you can replicate the option position by holding a dynamic hedge (this is basically what the black-scholes pde says).
hgibbs··on A Brief Explanation of Greg Egan’s Dust Theory
Well yes, you could construe any physical object as having conciousness by having a sufficiently detailed/complex schema that assigns 'meanings' to each of its states.

But we all agree a rock is not conscious, so this definition doesn't pass the common sense test.

One mistake is then claiming that this usage of the word conciousness can be substituted for the commonplace usage in all situations. E.g. why don't we feel the same ethical obligation to a potentially conscious rock that we do to animals, humans and, arguably, seemingly intelligent entities?

For a nice exposition of this one can read chapter 2 of Gary Drescher's book 'good and real'.

hgibbs··on Want people to do the right thing? Save them the guilt trip
Citations?
hgibbs··on The world should think better about catastrophic and existential risks
The definition from Wikipedia reads: "The black swan theory or theory of black swan events is a metaphor that describes an event that comes as a surprise, has a major effect, and is often inappropriately rationalised after the fact with the benefit of hindsight."

So predictability is not a criteria.

hgibbs··on Counterfactual Regret Minimization with Kuhn Poker
This is cool. I was actually asked for the optimal strategy for player 1 in this game as an interview question for a quant research role; I didn't know anything about game/decision theory, other than the basics, and so I only got as far as determining what to do with an Ace and a King, and then figuring out that you should bluff with a certain frequency if you hold the ace.

Anyhow, I came up with another question while considering this problem. Player 1 should only bluff at the optimal rate if player 2 punishes them for deviating from this rate by changing the frequency that they call when holding the king. Otherwise, Player 1 gains more on average by bluffing with the queen more frequently. So, how should Player 2 figure out if Player 1 is actually bluffing at the optimal rate (and not higher)? It could happen, with exponentially small probability, that Player 1 decides whether to bluff randomly, and bluffs with the first 10,000 queens before ever folding with a queen. How does Player 2 distinguish this from Player 1 only ever bluffing with the queen?

But perhaps this is something that the Nash Theorem deals with? I don't know. Any comments from people who know more than me on this are welcome!

hgibbs··on The Central Limit Theorem and Its Misuse
Interestingly, the O(1/sqrt(n)) error in the CLT provided by the Berry-Eseen Theorem is the best one can do for general sequences of our RVs. However, if you are happy to put certain 'bumps' in the normal distribution then you can improve the error to O(n^(-r/2)), where I think r is some integer depending on the smoothness of your random variables. These bumps are called an edegworth expansion.
hgibbs··on Books on Critical Thinking
Why is it unfortunate?
hgibbs··on Habits of Successful Language Learners
I've taken this approach while learning polish, which is the first language of my partner's parents. I started three years ago, and the first tactical decision I made was to take classes - not for the sake of being taught, but for the accountability that comes from having to regularly attend and do homework for a class. I also decided to not obsess over my progress or speed, and have been content taking a few years to learn what some people manage to do in six months of hard study. I figure it takes children 5-6 years to get any good, and they are effectively studying a language full time with seemingly much higher stakes (e.g. the necessity of communicating that you don't want broccoli for dinner).
hgibbs··on Aussie ISV buys ads to wake up Google support
Not all companies can afford in-house solutions.
hgibbs··on Why the French love to say no
"Yeah, nah" means one of two things. Most frequently it means "I agree with you that the answer is no", but it can also be used to say "yes I can see how that point is valid, but I disagree with the conclusion". Tone and context are used to differentiate between the two uses, particularly how one says the "nah" (in agreement or not). So for example: "it doesn't look like it's going to rain today aye" could be answered with a "yeah, nah" to indicate that one agrees that it won't rain today, or a "yeah, nah" to indicate that, while it may look like it won't rain, the respondent still believes it could rain.

I believe "nah, yeah" is used in an analogous way, but in my experience it's less commonly said than "yeah, nah".

hgibbs··on What is ergodicity? (2016)
Well, trivially, a dynamical system with that switches between two states is ergodic. We always know what the past state of the system was given the present.

No, ergodicity is more about visiting the each part of the phase space with some positive frequency i.e. on a long trajectory you see everything that can happen, and not too infrequently.

hgibbs··on Fuller House: Exposing high-end poker cheating devices (2016)
Incidentally, while having a look for Ruelle's book on thermodynamic formalism I stumbled upon a book title something along the lines of "the dynamics of roulette". I expected something about rare events in probability theory, perhaps betting strategies for optimising expected profit for certain levels of risk, but it ended up being on the mechanics of the roulette table itself - pages upon pages of hefty calculations of roulette balls including spin, velocity etc etc. It is crazy how much effort people put into getting an edge at a game.
hgibbs··on French economist scours spreadsheets to find secret offshore accounts
Without reading the article at all, this kind of thing sounds like an interesting mathematical problem.

You have some time series above various nodes on a graph (each node is an account), and you cannot observe the time series directly, only some functions of the total state space, such as yearly income statements, ( i.e. some information is hidden or obfuscated). The question is how much of the underlying graph and time series you can reconstruct from your observations. I know some people have methods for reconstructing time dependent dynamical systems from observations using E.g. manifold learning, but I'm not sure if this problem exactly fits the bill.

hgibbs··on Aerobic exercise is an effective treatment for depression
I actually notice my mood and energy levels drop significantly when I reduce my weekly exercise.
hgibbs··on I realized that derivatives are linear
Yes, you are right. My apologies.
hgibbs··on I realized that derivatives are linear
I think you'll find that derivatives are overwhelming defined as operators on vector spaces. It is definitely correct to talk about a linear function on a vector space.
hgibbs··on I realized that derivatives are linear
Depends on what space you define the derivative on. It is of course a bounded (and therefore continuous) operator from C^k to C^{k-1} for any positive integer K.

Additionally, it only really makes sense to talk about bounded operators between topological vectors spaces (as you need to make sense of what it means to be bounded), of which the most commonly dealt with are Banach spaces.

hgibbs··on Math problems that challenge the world
This article contains some glaring mistakes e.g. the statement of the Riemann hypothesis is wrong. I suspect the author tried to edit the mathematics for the sake of comprehensibility and, due to their unfamiliarity with the area, failed.
hgibbs··on Mathematical Notation Is Awful
The post is a pretty childish rant. One of the great facets of mathematical writing (and not learning, unfortunately) is that you can explicitly define your own notation, and then use said notation whenever you want. The author even notes this in his final paragraph, but doesn't seem to see it as an advantage of mathematical writing.

Prior to modern notation, mathematics was written out in english in full. What we have now is significantly better than what existed before modern mathematical shorthand.

Also, the expectation is an operator and not a 'function' (in the sense that it does not take values in one of the canonical scalar fields e.g. R, C). The notation makes perfect sense in this setting. For example, the expression E[x^2] should be interpreted as E acting on the function x -> x^2 and not on a number x.

hgibbs··on Biggest mystery in mathematics in limbo after cryptic meeting
There is literally only a handful of mathematicians that are qualified and they have gone to Oxford to sort it out - bar mochizuki.
hgibbs··on LSD Microdosing
Volumetric dosing. Put 100ug in 100ml of (distilled) water, measure 10ml and consume.
hgibbs··on Cutting a Klein Bottle in Half [video]
I had a really bad day. Cheers man you have no idea how much that means to me.
hgibbs··on Cutting a Klein Bottle in Half [video]
Funny story. I saw this video yesterday morning, about an hour before my topology exam. One of the questions involved calculating the Euler poincare characteristic of a 3 fold sum of RP2. Unfortunately, I had forgotten the formula for the case of m-fold sums of RP2 but I knew that it was linear in m and that the Euler poincare characteristic of RP2 was 1. Luckily, I remembered a Klein bottle is homeomorphic to the 2-fold sum of RP2. Having seen this video and knowing the cell complex structure of a mobius band ended up being enough for me to derive the answer.
hgibbs··on The Unreasonable Effectiveness of Recurrent Neural Networks
I did have a bit of a chuckle when they got to Algebraic Geometry. That's incredible.
← PreviousPage 2 of 3Next →