dia80

I'm sure your comment is written in excellent faith after some reflection. While i am unable to weigh in on the issue at hand I would offer a few cautionary scraps about luminaries of a field being able to accept it's wrong.

The Intelligence Trap[1] is an excellent book that among other things explains how being very clever can make it harder to get at the truth. Formidable intellects can patch any holes that appear in their world view by constructing convincing and credible arguements.

"Science advances on funeral at time" paraphrased from Max Planck - the individuals he was thinking of were truth seeking super brains as well.

[1] https://www.goodreads.com/book/show/41817546-the-intelligenc...

I live in South Hampstead about 3-4 miles from dead centre of London. We are very fortunate to have a garden. It is surrounded by many other gardens. Our building was constructed in the 1880s. There are many very old beautiful trees in the adjoining gardens, our garden has a forest feel at times. I'm filled with gratitude to some late Victorians who, 140 years ago, planted some small sapplings in their new back garden and probably didn't see them grow to anything very impressive.

As I understood it you would start choosing treatments at random, as one or more treatments started to outperform the others they are chosen with higher frequency. A sort of on-line trial. No AI involved.

I had wondered if a bandit algorithm would be an effective approach. Can anyone who knows more about it than me comment?

I don't think this is true. I trade cross currency basis and if there was excess demand for offshore USD the "basis" (premium over normal rates to borrow USD in return for lending e.g. EUR ) would be moving dramatically. It changed a bit today in response to USD shortage but this seems predominantly a domestic story.

That's unwise, you can call the US Bureau of Labor Statistics (BLS) up and ask them how they arrived at their numbers. In the UK you can go on field trips with the Office of National Statistics (ONS) to watch them measure inflation. It's materially different.

I don't think it's that bad in your garden... I ran a foobot at home when we had our son. Air quality in the house was good a mile or two from central london and a few hundred meters from a big road. Looking at the pollution maps the pollution is extremely concentrated along big roads and even a short distance (10s of meters) away it's not so bad.

The last time I had this feeling was when shorting groupon. The emperor has no clothes, I can't see how UBER ends up being materially different from a donut.

There isn't any informational value in constantly predicting crisis, of course you will be correct some of the time.

I am always amazed by Americans love of 'period' style architecture in preference to a more modernist style. Especially when I see friends in my age group (30s) pack there homes with dark wooden furniture that looks like it would be more at home in Versailles. To me it is just visually exhausting and I crave the relative calm of some minimalist sleek lines.

Limits to arbitrage, same reason GBTC the bitcoin trust trades at 100% premium.

It means they will allow their account at the exchange to settle client trades. This will involve them using their existing collateral at the exchange to underwrite the client trades. They will in turn demand collateral from the client but may be potentially exposed if the client defaults.

I recall it was Moscow state university they wouldn't admit any, other schools would

I recall reading elsewhere it was zero, they simply weren't allowed to admit Jewish students and these problems were just to try and cover up the discrimination.

I drive on London every day. Uber drivers are perpetually distracted. Either looking for a pickup or aimlessly meandering waiting for their next job. There are some serious negative externalities, much like the article.

A Bayesian mindset. When investigating a signal you have to ask: Is it too good to be true? Does it make economic sense?

Then simulate rolling optimisation and trading to see what returns could have been achieved with knowledge only available at the time.

If you don't persevere you achieve nothing. That fact alone is often enough to keep me going.

Deep torture, anyone? Gradient descent on stimuli to get what you want out of adversaries. Kind of scary

This would really help cushion the fiscal fallout from brexit for the rest of the UK.

I'm really pleased for this guy and hope he can continue to make progress.

Unfortunately the information content of these events is at best modest. After spinal cord injury you can get "spinal shock" which can last for a month or two. When it wears off you can recover a lot of previously lost function. As other comments point out bigger trials will be required to prove any efficacy.

Also there are many things in development to help acute spinal cord injury (some of which are FDA devices) but chronic SCI is a much thornier and challenging problem.

I'm guessing the tremendous amount of smoke and all the obstructions on the roof would have made this impractical. The situation wouldn't have been aided by helicopters falling on to rescuers below.

I just don't see the point of all these simple introductions to some ML topic. If you want to do a good job it is hard. If you need a simple introduction and don't just dig in a persevere with harder stuff you probably aren't going to succeed in whatever ML task you aspire to.

Thanks for this. If you wanted to get started doing this kind of analysis in python check out the examples for the maturing pyMC3

https://pymc-devs.github.io/pymc3/stochastic_volatility/

I think you mean any startup with a limited liability structure!

That works until it doesn't, see Japan.

This only possible because companies want to be able to fire their employees quickly so that means the employees can leave quickly. I know of one place (in the UK) with a one year notice term. You can't leave quickly so their business is secure but you are secure too as they have to give 12 months notice before letting you go.

Fixed income trading desk, large global hedge fund - London, UK (ONSITE,VISA)

Fixed income markets (bonds, futures, interest rate swaps) are going through big changes as electronic trading takes over. We are seeking a "Tech Lead" for our desk who can help us deploy new and existing strategies in emerging electronic trading venues now and recruit and lead our tech time through time.

Our desk of 3 has an informal and collegiate atmosphere. We have been setup around 2Y and have performed well however there is exciting potential for rapid tech driven growth. IMHO we have interesting problems and get to work with cool stuff e.g. machine learning

We will happily consider someone less experienced (2+ years commercial experience) if you can demonstrate excellence (solved hard problems in your PhD / great github profile / something else). We can sponsor a work visa for exceptional non-EU applicants.

Our production code is in python, C# and haskell but we are tech agnostic and will look to you to chose technologies as appropriate.

Requirements:

  * Probably CS PhD (haven't seen anyone up to scratch with out this so far but since this is HN I know I could here.)
  * Familiar with python 
  * Minimum 2y commercial experience
  * Knowledge of good practices - e.g. git, unit testing

Compensation: Excellent (no joke) and potential for rapid growth as we perform

Please send your CV to the email in my profile.

Thanks for your attention and hope to hear from you,

David

It's a bit more complex than the Greek's have spent too much money and expecting Germans to bail them out. This situation was an inevitability [1] given the Eurozone's make up and the political classes being in denial has worsened it.

[1] http://blog.mpettis.com/2015/02/syriza-and-the-french-indemn...

London | Large Hedge Fund | ONSITE

My 3 person interest rate trading group is seeking a talented developer with MSc or above CS qualification who can help us grow through automating workflow and implementing new trading models.

We have production code in python, C# and haskell.

You will need experience of python and be able to demonstrate that you have solved tough problems in the past (academic / work / personal projects).

If you are motivated, smart & proactive with strong quantitative skills we can provide an environment where your contribution will make an impact on the bottom line and you will be rewarded appropriately.

Great comp + benefits

Drop me an email with your CV (in my profile) Thanks!

This kind of concerns me. I have a T4/T5 spinal cord injury from a snowboarding accident 4.5Y ago. I have good hand and arm function but that's it no abdominals etc. I drive myself to work everyday and transfer in and out and fold up my wheelchair.

London's black cabs all have ramps. UBER is making big inroads here and I'm worried that eventually it'll be tricky to get a black cab. I can transfer into a regular car but I would need the driver to put my wheelchair in the boot (trunk) and given busy streets, harassed drivers, needing to pull right up to the kerb etc. I've not really had the guts to try it. It's particularly relevant here as the tube is mostly inaccessible and the buses painfully slow.