WELCOME TO ZALANDO RESEARCH

SCALING TECH IN FASHION WITH THE POWER OF EXPERIMENTATION IN THEORY AND PRACTICE.

About

Zalando Research is an endeavor to place Zalando at the forefront of cutting-edge research and to complement our already strong foothold on technology. We want to carve out an academic starting point for Zalando Technology. We aim to give our great researchers an outlet where they can be better organized and aligned. Finally, also want to have a greater impact in the tech research community, especially in the fields of data science, machine learning, and artificial intelligence.

Mission

We believe research should not be enclosed in the ivory towers of abstract and theoretical hypotheses. Our mission is to conduct research that is derived from real customer problems and challenges we face in our everyday work. Putting the customer at the forefront of our thinking is paramount for Zalando Research.

Approach

We enable our research scientists to proactively exchange knowledge with others in the research communities. On top of speaking at conferences and publishing papers, we empower research collaborations with the global industry and academic partners. Our goal with Zalando Research is to have an official mandate for our talented scientists to work from, which also insists on collaborative projects within the greater academic community.

Projects

Publications

Probabilistic Time Series Forecasting with Implicit Quantile Networks

Adèle Gouttes , Kashif Rasul , Mateusz Koren , Johannes Stephan , Tofigh Naghibi

ICML 2021 Time Series Workshop

July, 2021

Details PDF Code

Grid Partitioned Attention: Efficient TransformerApproximation with Inductive Bias for High Resolution Detail Generation

Nikolay Jetchev , Gökhan Yildirim , Christian Bracher , Roland Vollgraf

under review

July, 2021

Details PDF Code

Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting

Kashif Rasul , Calvin Seward , , Roland Vollgraf

ICML 2021 Conference

February, 2021

Details PDF Code

A Rigorous Theory of Conditional Mean Embeddings

Ilja Klebanov , , T. J. Sullivan

SIAM Journal on Mathematics of Data Science

September, 2020

Details PDF

Markov Chain Importance Sampling – a highly efficient estimator for MCMC

, Ilja Klebanov

Journal of Computational and Graphical Statistics

September, 2020

Details PDF

More Publications

Blog

Forecasting time-series and their interactions

In this blog post we present our novel probabilistic multivariate time series prediction model which learns the temporal distribution of multivariate data via conditioned normalizing flows.