From the 1 of 9 linked papers with an AI index.
9 papers
NMINE: Normalized Mutual Information Neural Estimation
Petra Eerikinharju, Marko Tuononen, Ville Hautamäki
The paper introduces a fully neural estimator for normalized mutual information of continuous, multidimensional variables, combining a MINE-based mutual information estimator with…
How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection
Ivan Kukanov, Janne Laakkonen, Ville Hautamäki
Meta-learning for domain generalization (MLDG) improves out-of-distribution speech deepfake detection over empirical risk minimization (ERM) when both objectives train low-rank ada…
Zero-Shot Heart Rate Variability Forecasting from Consumer Wearables Using Time Series Foundation Models
Luukas Peräkylä, Fahad Sohrab, Ville Hautamäki +3
Short-term Heart Rate Variability (HRV) forecasting could provide clinicians with actionable lead time for detecting autonomic dysfunction and adverse cardiac events. Consumer wear…
Targeted Fine-Tuning of DNN-Based Receivers via Influence Functions
Marko Tuononen, Heikki Penttinen, Ville Hautamäki
We present the first use of influence functions for deep learning-based wireless receivers. Applied to DeepRx, a fully convolutional receiver, influence analysis reveals which trai…
Zero-shot World Models via Search in Memory
Federico Malato, Ville Hautamäki
World Models have vastly permeated the field of Reinforcement Learning. Their ability to model the transition dynamics of an environment have greatly improved sample efficiency in…
Mixture of Low-Rank Adapter Experts in Generalizable Audio Deepfake Detection
Janne Laakkonen, Ivan Kukanov, Ville Hautamäki
Foundation models such as Wav2Vec2 excel at representation learning in speech tasks, including audio deepfake detection. However, after being fine-tuned on a fixed set of bonafide…