Data efficiency
Numerous applications of Machine Learning in business intelligence, process optimization, product development or sales do not benefit from true “big data”. Blind application of state-of-the-art methods can lead to catastrophic failure. Proper data acquisition and modelling are key to successful applications.

All content in this series

Pill

Resolving Training Biases via Influence-based Data Relabeling

Influence functions are used to correct corrupt labels in a dataset that would significantly decrease a trained model’s performance. …

A photo of Fabio
Data Efficiency
Seminar

An introduction to Unsupervised Neural Machine Translation

Lack of large parallel corpora or dictionaries for most language pairs and modern data-hungry models make the need for unsupervised …

Sebastian Wagner
Data Efficiency
Seminar

Dealing with imbalanced classes

In this talk Wolfgang shares some of his personal experience and methods while training classifier CNNs on image datasets with highly …

Wolfgang Reuter
Data Efficiency
Seminar

Neural Network Dynamics for Model-Based Deep Reinforcement Learning with Model-Free Fine-Tuning

Paper abstract: Model-free deep reinforcement learning algorithms have been shown to be capable of learning a wide range of robotic skills, …

Fabian Hainzl
Data Efficiency
Efficient RL
Seminar

A short introduction to model-based Reinforcement Learning

Model-based RL learns a deterministic or probabilistic forward model of the dynamics of an environment for action planning or direct policy …

A picture of Miguel
Data Efficiency
Efficient RL
Seminar

An introduction to data distillation

Distillation ensembles predictions from multiple transformations of unlabeled data, using a single model, to automatically generate new …

Ishmeet Kaur
Data Efficiency

Other series in Efficient Machine Learning

Data valuation

Attributions of value to training samples can be used to examine data, improve data acquisition, debug and improve models or compensate providers. Recent developments in the field enable principled and useful definitions of value which …

10 items

Representation Learning

Representation learning is concerned with finding semantically useful and compact, i.e., low dimensional, representations for inputs from selected data distributions. These representations can be efficiently used for downstream tasks, …

1 item

Efficient RL

Recent and popular advances in Reinforcement Learning are known to be data-hungry. Attempts to handle this deficit include developing complex simulators while improving the sim2real transition of models, explicitly modelling the …

9 items
Check all of our work