flashfaffe2

Thanks for the refreshing reminder. This was one of my favourite topic at school but ended never used it in professional environment. However , I will definitely review the theory. Banach spaces and fix theorem were kind useful definition/tool but I always struggled to understand practical applications.

Thanks for sharing the article

To the best of my knowledge, most of the founding team started their careers in the US ( meta,etc..) and their primary investors are US VCs. In that regard, they smartly benefit on both side : US funding and European brains

Without being to self-centred here but since I have been using LLM heavily I have always challenged the results given

The post seems to propose the following vector:

Idea-> LLm validation -> confidence -> no further checks

My process is more :

Idea-> LLm response -> skeptical reflection -> adversarial prompting -> synthesis

Thanks for the inputs.

"Bandits work best for relatively simple optimization choices at very large scale"

Have you considered different methods to address this shortcoming?

Genuine question for an outsider: would this imply that gold can be created? My memories from the last chemistry class I had, I clearly remember my teacher demonstrating philosopher stone ( aka changing materials in gold ) was feasible.

To resonate with recent article published on YC, the era of real estate is over.

Would be interested to know how this was computed. I guess this was a kind of survey and if so, based on the question: are proud of the US army or are you supporting us army action around the world then I wouldn't be surprised to observe difference.

Prévois thread on Spinoza here: https://news.ycombinator.com/item?id=37150843

Asset management business model in needs of clear reassement. There is clearly a lack of concurrence combined with high fees. Moreover, the sector has in a recent years seen an inflation of legislative obligation impacting their costs.

Short debate on the same topic on french TV:

https://youtu.be/5Sf6hdnSqS4

I only know Cramer from the fun and others meme is been the originator. But that statistics should be taken with caution. Being long equity since 2008 was not really a proof of creating a alpha ( aka adding superior return than the market). That's the long term low interest rates environment that incentivize W Buffet to make his famous 10 years long sp500 won't be beaten by any HF.

The current system requires more workers ( with decent salary) than number of retirers... Given the population if getting older this creates an a very prévisible imbalance...

What I personally did but very hard to convince others to follow that way...

French here:

Though the problem is well known (at least for 40 years), No french governments has considered diversifying the source of the revenue funding the pension system. Actually its worse as the only capitalisation existing are in a process to be nationalised.

Moreover, this reform is purely the outcome of political game. Macron electorat is mainly the 'boomers' who fully beneficiated of th situation ( in fact their pension are impacted despite having higher wealth than the majority of the workers).

I personally would consider a reform only if there an effort requested for everyone.

The withdraw from the $59 trillion Net Zero Asset Managers initiative shows somehow a little more nuance than the current doxa. Like it or not, this echoes the comment from former HSBC manager fired few months back for raising his own view.

Same author but More technical video: https://youtu.be/IlliqYiRhMU

TEDx video from Ramin hasani explaining his algorithm.

https://www.youtube.com/watch?v=RI35E5ewBuI

Low Interest-->Crony capitalism in a real free market loss would have occurred by the accountable people.

I totally second this. For someone who has a tortuous relationship with mathematics, I always lovedthe beauty of som concepts but to a certain point, got lost mainly due to knowledge base not really understood. Getting to know some historical facts would have help to be more critical to myself ( rather than just copy and paste formula).

Years later I surprised myself spending time on new maths problems thanks to YouTube maths contents(3brown1blue, Michael Penn, mindyourdecisions,etc...) Which make me reboot my knowledge. The challenge here is how is it inseminated during your scholarship.

From the what I remember, complex number theory were introduced to me as a rule that need to be acknowledged ( with no incentive to look further).

Welsh lab vid below made me reconsider my own shortcomings.

https://youtube.com/playlist?list=PLiaHhY2iBX9g6KIvZ_703G3KJ...

Exactly what I thought reading the comment.

Maybe this is off topic but how this algorithm adapt itself to time series?

I mean you want to cluster data which might be autocorrelated... Does this case is taken into account or do you need to preprocess your data beforehand?

Indeed, this seems worth a deep read as this especially address main PCA shortcomings ( heterogeneous data, non numerical data,.etc...). Thanks mate I've definitely find a way to keep myself busy this weekend.

Not sure I got your point. Can you elaborate?

Kind of contrarian view I like to read on HN. If I may add, maybe the reason people used robo advisory here is potentially ( mainly) because they are not aware of the other tool.

I must admit myself that the focus over fees charged is not something, as new comers, you would look at priority. Especially as sales person in the sector would not put that out.

As customer who experienced this not far than yesterday, my priority was to flight asap. Not to enter in litigation with the company ( which is incredibly painful). Moreover if you pay your ticket peanut ( Ryanair or easyJet in Europe), I doubt you will consolate yourself with that advancement.

Agreed with all above. I would add that this won't help customers either as regardless you paid upon check-in or not, this does not address the ultimate point for customers: go to their destination on agreed time.