4 papers
Semi-overlapping Multi-bandit Best Arm Identification for Sequential Support Network Learning
András Antos, András Millinghoffer, Péter Antal
Many modern AI and ML problems require evaluating partners' contributions through shared yet asymmetric, computationally intensive processes and the simultaneous selection of the m…
Characterization of Transfer Using Multi-task Learning Curves
András Millinghoffer, Bence Bolgár, Péter Antal
Transfer effects manifest themselves both during training using a fixed data set and in inductive inference using accumulating data. We hypothesize that perturbing the data set by…
DTI-GP: Bayesian operations for drug-target interactions using deep kernel Gaussian processes
Bence Bolgár, András Millinghoffer, Péter Antal
Precise probabilistic information about drug-target interaction (DTI) predictions is vital for understanding limitations and boosting predictive performance. Gaussian processes (GP…
BandiK: Efficient Multi-Task Decomposition Using a Multi-Bandit Framework
András Millinghoffer, András Formanek, András Antos +1
The challenge of effectively transferring knowledge across multiple tasks is of critical importance and is also present in downstream tasks with foundation models. However, the nat…