Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

TimesFM: Time Series Foundation Model for time-series forecasting

github.com

Post reply on HN
Print view View original

71–80 of 123 posts

Re: TimesFM: Time Series Foundation Model for time-series forecasting

#71

Post by bdjsiqoocwk »

itronitron wrote: post #68
whimsicalism wrote: post #17

I'm curious why we seem convinced that this is a task that is possible or something worthy of investigation. I've worked on language models since 2018, even then it was obvious why language was a useful and transferable task. I do not at all feel the same way about general univariate time series that could have any underlying process.

There was a paper written a while back that proved mathematically how you can correlate any time series with any other time series, thus vaporizing any perception of value gained by correlating time series (at least for those people that read the paper.) just wanted to share

What does that mean "you can correlate"? That phrase is meaningless.

Re: TimesFM: Time Series Foundation Model for time-series forecasting

#72

Post by notnaut »

itronitron wrote: post #68
whimsicalism wrote: post #17

I'm curious why we seem convinced that this is a task that is possible or something worthy of investigation. I've worked on language models since 2018, even then it was obvious why language was a useful and transferable task. I do not at all feel the same way about general univariate time series that could have any underlying process.

There was a paper written a while back that proved mathematically how you can correlate any time series with any other time series, thus vaporizing any perception of value gained by correlating time series (at least for those people that read the paper.) just wanted to share

I would like to read more. Feels sort of like an expression of certain “universal truths” like the 80/20 rule or golden ratio

Re: TimesFM: Time Series Foundation Model for time-series forecasting

#73

Post by ukuina »

aeyes wrote: post #31
tasty_freeze wrote: post #22

Earlier quoted context omitted.

Game developers are constantly trying to minimize lag. I have no idea if computers are so fast these days that it is a "solved" problem, but I knew a game developer ages ago who used a predictive mouse model to reduce the apparent lag by guessing where the mouse would be at the time the frame was displayed (considering it took 30 ms or whatever to render the frame).

Quake internet play only became acceptable when client side prediction was implemented, I'm sure it would be better to have real prediction instead of simple interpolation. https://raw.githubusercontent.com/ESWAT/john-carmack-plan-ar...

What an amazing look into one of the greatest minds in programming!

Thank you for this treasure.

The relevant bits:

> I am now allowing the client to guess at the results of the users movement until the authoritative response from the server comes through. This is a biiiig architectural change. The client now needs to know about solidity of objects, friction, gravity, etc. I am sad to see the elegent client-as-terminal setup go away, but I am practical above idealistic.

> The server is still the final word, so the client is allways repredicting it's movement based off of the last known good message from the server.

Re: TimesFM: Time Series Foundation Model for time-series forecasting

#75

Post by zeroxfe »

whimsicalism wrote: post #63
zeroxfe wrote: post #51

Earlier quoted context omitted.

> I'm curious why we seem convinced that this is a task that is possible or something worthy of investigation. There's a huge industry around time series forecasting used for all kinds of things like engineering, finance, climate science, etc. and many of the modern ones incorporate some kind of machine learning because they deal with very high dimensional data. Given the very surprising success of LLMs in non-langua…

Task specific time series models, not time series “foundation models” - we are discussing different things.

I don't think we are. The premise of this is that the foundation model can learn some kind of baseline ability to reason about forecasting, that is generalizable across different domains (each which needs fine tuning.) I don't know if it will find anything, but LLMs totally surprised us, and this kind of thing seems totally worthy of investigation.

Re: TimesFM: Time Series Foundation Model for time-series forecasting

#76

Post by viraptor »

I'm not sure I understand two things here. Could someone clarify: 1. This is a foundation model, so you're expected to fine tune for your use case, right? (But readme doesn't mention tuning) 2. When submitting two series, do they impact each other in predictions?

Re: TimesFM: Time Series Foundation Model for time-series forecasting

#77

Post by viraptor »

morkalork wrote: post #35
leeoniya wrote: post #23

Earlier quoted context omitted.

LSTM is common for this. also https://facebook.github.io/prophet/

How data hungry is it, or what is the minimum volume of data needed before its worth investigating?

The more complex the data is, the more you need. If your values are always 5, then you need only one data point.

Re: TimesFM: Time Series Foundation Model for time-series forecasting

#79

Post by refibrillator »

whimsicalism wrote: post #17

I'm curious why we seem convinced that this is a task that is possible or something worthy of investigation. I've worked on language models since 2018, even then it was obvious why language was a useful and transferable task. I do not at all feel the same way about general univariate time series that could have any underlying process.

Why do you think language is so special?

There's an extensive body of literature across numerous domains that demonstrates the benefits of Multi-Task Learning (MTL). Actually I have a whole folder of research papers on this topic, here's one of the earliest references on hand that I feel captures the idea succinctly in the context of modern ML:

“MTL improves generalization by leveraging the domain-specific information contained in the training signals of related tasks" [Caruana, 1998]

I see repetition and structure everywhere in life. To me it's not far fetched that a model trained on daily or yearly trends could leverage that information in the context of e.g. biological signals which are influenced by circadian rhythm etc.

Disclaimer: my background is in ML & bio-signals, I work with time series too much.

Re: TimesFM: Time Series Foundation Model for time-series forecasting

#80

Post by jimmySixDOF »

itronitron wrote: post #68
whimsicalism wrote: post #17

I'm curious why we seem convinced that this is a task that is possible or something worthy of investigation. I've worked on language models since 2018, even then it was obvious why language was a useful and transferable task. I do not at all feel the same way about general univariate time series that could have any underlying process.

There was a paper written a while back that proved mathematically how you can correlate any time series with any other time series, thus vaporizing any perception of value gained by correlating time series (at least for those people that read the paper.) just wanted to share

The only other timeseries paper I am aware of is TimeGPT

https://news.ycombinator.com/item?id=37874891

Post reply on HN