Gareth Davies

I research machine learning for applications in time series forecasting and tabular data predictions

[Gareth Davies @garethmd]

https://t.co/JatQxPrxV5

[Gareth Davies @garethmd]

https://t.co/kLfgd3MY1M

[Gareth Davies @garethmd]

Revin: What really grinds my gears https://t.co/ncz9mAdbwU

[Gareth Davies @garethmd]

NeurIPs Ariel Challenge: What's it like to take part in a Kaggle competition? - Neural Aspect https://t.co/cqVxYIm48j

[Gareth Davies @garethmd]

Multivariate Time Series models: Do we really need them? - Neural Aspect https://t.co/yxoHWm36Yp

[Gareth Davies @garethmd]

Check out my post on LTSF-Linear: Embarrassingly simple #timeseries forecasting models. Training with channel independence is somewhat tedious. I've done the hard work so you don't have to. https://t.co/Wpa6xfHAA9

[Gareth Davies @garethmd]

Allez Allez. Can MXNet still keep pace with the rest of the pack? Find out in my benchmark of train time compute across 40 datasets with Pytorch. https://t.co/BQz8knuSQ6

[Gareth Davies @garethmd]

We launch our Formula 1 themed #TimeSeries Model Championships. So head on over to the starting grid at https://t.co/Muc6KGIniC to find out more.

[Gareth Davies @garethmd]

I don't care what #GenerativeAI is and you shouldn't either - Neural Aspect https://t.co/57jhSGFJIh

[Gareth Davies @garethmd]

Take a look at my post on how to get state of the art #TimeSeries #forecasting results using machine learning with my variant of DeepAR and 1Cycle Scheduling @lnsmith613 . https://t.co/fiCEmRGlP3

[Gareth Davies @garethmd]

Want to know how to get a performance boost from your neural network for virtually no extra effort? Take a look at my review of Leslie N Smith's @lnsmith613 paper on super-convergence. https://t.co/7nntCbAF3q

[Gareth Davies @garethmd]

Supercharge your Neural Networks with Super-convergence - Neural Aspect https://t.co/7nntCbAF3q

[Gareth Davies @garethmd]

A fistful of MASE: Deconstructing DeepAR - neural aspect https://t.co/Ov5bHf7347

[Gareth Davies @garethmd]

Excited to share my first paper on arXiv!
Our research delves into Autoregressive Recurrent Neural Networks for time-series forecasting, shedding light on their limitations with future, time-dependent covariates or leading indicators. Enjoy!

https://t.co/pLLlTLYiMR

[Gareth Davies @garethmd]

Is it worth predicting covariates in an LSTM for timeseries forecasting? - neural aspect https://t.co/YAMk61LskR

[Gareth Davies @garethmd]

@krasul @NielsRogge @huggingface perfect yes I'll do that.

[Gareth Davies @garethmd]

CloudApps get accredited by GRI http://t.co/YX1Z3JLN

[Martin Fowler @martinfowler]

posted: an infodeck exploring how to deal with developing software for multiple mobile devices http://t.co/Bip1RJCL

[Gareth Davies @garethmd]

We're hiring http://t.co/bpGlKB1P developers @CloudAppsCarbon. Apex experience a must. http://t.co/sCgap9tj