Deep Learning

1LoRA: Summation Compression for Very Low-Rank Adaptation

We propose 1LoRA (Summation Low-Rank Adaptation), a compute, parameter and memory efficient fine-tuning method which uses the feature sum as fixed compression and a single trainable vector as decompression.

Alessio Quercia, Zhuo Cao, Arya Bangun, Richard D. Paul, Abigail Morrison, Ira Assent, Hanno Scharr

Enhancing Monocular Depth Estimation with Multi-Source Auxiliary Tasks

We propose to improve MDE while using auxiliary datasets from related vision tasks with an alternating training scheme with a shared decoder built on top of a pre-trained vision foundation model.

Alessio Quercia, Erenus Yildiz, Zhuo Cao, Kai Krajsek, Abigail Morrison, Ira Assent, Hanno Scharr

MRI Reconstruction with Regularized 3D Diffusion Model (R3DM)

We propose a 3D MRI reconstruction method that leverages a regularized 3D diffusion model combined with optimization method.

Arya Bangun*, Zhuo Cao*, Alessio Quercia*, Hanno Scharr, Elizabeth Pfaehler

Parameter-efficient Bayesian Neural Networks for Uncertainty-aware Depth Estimation

In this work, we investigate the suitability of PEFT methods for subspace Bayesian inference in large-scale Transformer-based vision models.

Richard D. Paul*, Alessio Quercia*, Vincent Fortuin, Katharina Nöh, Hanno Scharr

Parameter-efficient Bayesian Neural Networks for Uncertainty-aware Depth Estimation

SGD Biased towards Early Important Samples for Efficient Training

Biasing SGD towards samples found to be more important early in training to improve training efficiency.

Alessio Quercia, Abigail Morrison, Hanno Scharr, Ira Assent

SGD Biased towards Early Important Samples for Efficient Training

Preictal onset detection through unsupervised clustering for epileptic seizure prediction

Unsupervised method to automatically detect interictal and preitcal labels from EEG signals for Epileptic Seizure Prediction.

Alessio Quercia, Thomas Frick, Fabian Egli, Nicholas Pullen, Isabelle Dupanloup, Jianbin Tang, Umar Asif, Stefan Harrer, Thomas Brunschwiler

Preictal onset detection through unsupervised clustering for epileptic seizure prediction

Epileptic seizure prediction

Epilepsy is a common neurological disorder characterized by recurrent epileptic seizures. These seizures have different intensities and might lead to accidents or, in the worst case, to sudden death. Therefore, being able to predict epileptic seizures would allow patients to be prepared, reducing the risk of injury.

Epileptic seizure prediction