Tim Januschowski

@tomaspfister @MergenthalerMax Last time around in March, Max gave the cohort a preview of Forecasting, Principles and Practice, the Pythonic way – I wonder what’s cooking this time! The course starts October 6, link for sign-up is in comments. This is not related to my work at Databricks.

For me, the outside impulses, be it via @tomaspfister this year or @MergenthalerMax et al in the deep dive are always the special sessions. Especially with #Nixtla, it’s great to have witnessed their development over the last few years.

Never change a winning team – and so it is with great pleasure to announce that our friends at #Nixtla will again return for the Q3 edition of “Modern Forecasting in Practice” with an in-depth one hour session focussed on the #Nixtlaverse.

We get started at 1pm, looking forward to meeting everyone! #ISFConf2024 @timoschowski @lostella @krasul

@predict_addict @confusionmatrix Maven yes, ISF not unfortunately 😕

@timoschowski @confusionmatrix Is there online version like in previous years?

Looking forward to this and grateful to be part of the learner crew. While the below is for academics, @confusionmatrix and me continue with our #forecasting course for industrial folks: maven.com/tim-januschows…

Happy to be featured in this campaign - @confusionmatrix and my next next cohort is in October and yes, genAI models for forecasting are going to feature but so is progress on forecast evaluation. #forecasting maven.com/tim-januschows…

***Internship Opportunity*** We're hiring interns. Come, join us in building an ML platform reinvented for real-time. You'll gain hands-on experience in building ML systems from the ground up. 💸 Stipend: ₹1L/month 🗓️ Start Date: May/June 2024 📍 Location: Virtual 🚀 Career Path: We'll roll out PPOs to top performers Open Roles: 1. ML 2. Infra 3. DevOps 4. Streaming Systems 5. UI/UX If you're interested, please fill out this form: forms.gle/SQxiXALXzztQyW… #internship #hiring

@predict_addict Thanks for the shout out. I can at best claim to have participated in the beginnings here (alongside the great @confusionmatrix ) - but the team has been hammering out things like crazy @syamasundar @LoVVgE @lostella @solitarypenman

I have been saying for years that Amazon Science has by far the best forecasting R&D team on the planet and what drives the innovation at the team is not focus on specific models but the methodology and in depth knowledge of forecasting which is becoming somewhat a lost art elsewhere. This is reflected in the results - whilst time series LLMs have a long way to go, out of the recent high profile LLMs Amazon Chronos whilst shown to underperform to stats ensembles in some cases it is an example of stellar work and on a relative basis is much higher quality LLM compared to alternatives like Lag Llama and Saleforce Moirai as shown independent tests. More importantly is Amazon’s commitment to transparency and open source compared to say Google which still didn’t release its code. What is even more interesting is integration of Amazon Chronos into AutoGluonTS with Chronos becoming just another model that users can reliably test on own data. This move from Amazon Science is a stroke of genius in my view, a masterpiece signifying the way one should be looking at forecasting models. No model is going to be the best in all scenarios but some including Chronos will be. And this is the only right way to look at forecasting and Amazon forecasting R&D team understands it well, that’s why they are the best forecasting R&D team on the planet, because they way how they look at time series planet 🌍 and innovate. Kudos to all great amazon and ex amazon forecasting folks like @timoschowski , @canerturkmen and many others for moving the field forward. @AmazonScience @autogluon #timeseries #forecasting

🚀 Excited to share that we released Chronos today. Chronos is a framework for building pretrained time series models based on language model architectures. Simple idea: quantize time series into tokens and feed them into 🤗 @huggingface models. 🧵

Full house for the tutorial on Time Series Forecasting with Deep Probabilistic Approaches @ECMLPKDD. Nicely presented by @timoschowski @LoVVgE and Jan Gasthaus #machinelearning #timeseries #deeplearning #forecasting

Snapshot of some happy folks at the IIF sponsored workshop @MonashBusiness in Prato! @robjhyndman leading a 3day event on Forecast Reconciliation robjhyndman.com/prato2023/inde…

If you'd like to attend, DM me -- include your affiliation and email address so that I can register you for the event and you'll receive more details then.

Monday, 11 September, 10:15am, @robjhyndman will give a talk on "Forecast Reconciliation", a topic that he's pioneered over the last decade. The talk will be in a #zalando building close to Ostbahnhof (not the main Zalando campus) and attendance is open to Zalando externals.

If you're a PhD student in ML, Econ or Operations Research, consider joining us for a 6 month research internship in #Zalando's Pricing Platform. You'll be working with folks like @patrickdoupe and myself. Details in the link: jobs.zalando.com/en/jobs/508468…

@IIForecasters @Google Honored to be part of it!

ISF 2023 is pleased to announce our full Practitioner Track line-up of speakers and sessions, coming to you courtesy of our sponsor @Google Thank you so much! #PRACTSPEAKERS" target="_blank" rel="nofollow noopener">isf.forecasters.org/program/speake… #forecasting #ISFConf2023

@armin_kekic @MonesRaslan Great to have worked on this together!

With former teammates Manuel Kunz, Stefan Birr, @MonesRaslan, Lei Ma, Zhen Li, Adèle Gouttes, Mateusz Koren, Tofigh Naghibi, Johannes Stephan, Mariia Bulycheva, Matthias Grzeschik, Michael Narodovitch, Kashif Rasul, Julian Sieber, @timoschowski. 10/

Glad we can finally share this publicly: new paper on how @ZalandoTech uses deep-learning based forecasting for algorithmic pricing. A rare insight into how the machine learning is used to solve real-world problems. 🧵 1/ 📃: bit.ly/3qrYZXj

@LoVVgE @adad8m @smolix @danielle_maddix @confusionmatrix This paper aged well!

@adad8m If I'm not mistaken, this plot is from our ICML paper? (cf. Figure 1 in proceedings.mlr.press/v97/wang19k/wa…) We actually used Powerpoint, and indeed took us less than 5 mins. 🙂

@RMINT_Balaji @get_sphere Thanks for having joined @confusionmatrix and me! Our next cohort is planned for October: maven.com/tim-januschows…

Just completed a course on Modern Forecasting in Practice through @get_sphere. Special thanks to Jan Gasthaus and @timoschowski for sharing their extensive knowledge and experience. More details here: getsphere.com/cohorts/modern… #forecasting #onlinecourse #ml #datascience