Eingabe löschen






We invite Data Science experts to present and discuss topics relevant to the scope of Datalab. The seminar usually takes place every two weeks, on Wednesdays from 12:00-13:00 at the Winterthur Campus of ZHAW.

Future Events

Date Time&Place Speaker Title Abstract
Wed, 23.6. 12:00-13:00, Zoom Lukas Tuggener (ZHAW CAI) Is it Enough to Optimize CNN Architectures on ImageNet? An implicit but pervasive hypothesis of modern computer vision research is that convolutional neural network (CNN) architectures that perform better on ImageNet will also perform better on other vision datasets. We challenge this hypothesis through an extensive empirical study for which we train 500 sampled CNN architectures on ImageNet as well as 8 other image classification datasets from a wide array of application domains. The relationship between architecture and performance varies wildly, between the datasets. For some of the datasets, the performance correlation with ImageNet is even negative. Clearly, it is not enough to optimize architectures solely for ImageNet when aiming for progress that is relevant for all applications. We identify two dataset-specific performance indicators: the cumulative width across layers as well as the total depth of the network. Lastly, we show that the range of dataset variability covered by ImageNet can be significantly extended by adding ImageNet subsets restricted to fewer classes.

Past Events (2020)

Date Speaker Title Slides
Wed 02.12. Pia Viviani
Citizen Science: What is it exactly and what are the challenges? PDF
Wed 21.10. Daniel Neururer
On the Importance of time dependent features for Speaker Recognition PDF
Wed 09.09. Raphael Weber, Mehmet Yesil, Prof. Dr. Ruedi Füchslin and Prof. Dr. Kurt Stockinger
Quantum Databases and Quantum Machine Learning – How Far Can We Go on a Publicly Available Quantum Computer? PDF
Wed 08.04. Dr. Ricardo Chavarriaga
Presentation of the CLAIRE Office Switzerland PDF
Wed, 11.03. Nicolas Schmid
Data-Driven Adaptive Controller Parameterisation: A Bayesian Optimization Approach PDF
Thu, 05.03. Prof. Philipp Hungerländer
(Univ. Klagenfurt)
Propagation of vibroacoustic quality specifications in the automotive industry PDF
Wed, 26.02. Ursin Brunner and Prof. Kurt Stockinger
Entity Matching with Transformer Architectures - A Step Forward in Data Integration Prezi
Wed, 29.01. Prof. Kerrie Mengersen
(Queensland University of Technology QUT)
Bayesian Statistics in the Big Data Era  

Past Events (2019)

Date Speaker Title Slides
Wed, 18.12. Prof. Dr. Luca Maria Gambardella
(Director of IDSIA, USI-SUPSI, Lugano)
Innovation through Artificial Intelligence PDF
Fri, 15.11. PD Dr. Friedhelm Schwenker
(University of Ulm)
Artificial deep neural networks for multimodal emotion recognition PDF
Thu, 12.09. Dr. Siavash Arjomand Bigdeli
Deep Density Approximation for Bayesian Image Restoration  
Fri, 09.08. Prof. Haiwang Cao
(Zhengzhou University of Aeronautics, China)
AI application in safety support for civil airports  
Mon, 15.07. Prof. Dr. Mark Sanderson
(RMIT University, Melbourne)
Conferences vs. journals in computer science  
Mon, 20.05. Dr. Vasileios Trigonakis
(Oracle Labs Switzerland)
Introduction to Graph Processing (with PGX)