Projektbeschreibung
Deep Learning auf die Metaebene bringen
Deep Learning ist ein Teilgebiet des maschinellen Lernens, das Computer lehrt, Schlussfolgerungen auf dieselbe Weise wie Menschen zu ziehen. Dazu werden kontinuierlich Daten analysiert. Das Gebiet hat bemerkenswerte Fortschritte im maschinellen Sehen, in der Spracherkennung, Verarbeitung natürlicher Sprache und im Verstärkungslernen zu verzeichnen. Das EU-finanzierte Projekt DeepLearning 2.0 verfolgt nun das Ziel, Deep Learning auf die nächste Stufe zu bringen. Es wird Lernmethoden auf Metaebene vorschlagen, welche die Ergebnisse und die Leistungsfähigkeit von Lernalgorithmen verbessern. Die neuen Methoden werden zur Erstellung neuartiger, maßgeschneiderter Deep-Learning-Pipelines ohne handgefertigte Elemente beitragen, die genauer und einfacher anzuwenden sowie in weniger Zeit trainierbar sind. Um die Realisierbarkeit der vorgeschlagenen Methoden zu demonstrieren, werden die Forschenden die neuen, individuell angepassten Deep-Learning-Pipelines für die Elektroenzephalographie-Dekodierung und RNS-Faltung implementieren.
Ziel
Deep learning has revolutionized many fields, such as computer vision, speech recognition, natural language processing, and reinforcement learning. This success is based on replacing domain-specific hand-crafted features with features that are learned for the particular task at hand. The logical step to take deep learning to the next level is to also (meta-)learn other hand-crafted elements of the deep learning pipeline. We therefore propose to develop meta-level learning methods for the creation of novel customized deep learning pipelines, by means of:
1. Hierarchical neural architecture searchfor learning qualitatively new architectures and architectural building blocks from scratch;
2. Learning of optimizers and hyperparameter adaptation policies that adapt totheir context in order to converge faster and more robustly;
3. Learning the data to train on, to remove the need for large sets of labelled data; and
4. Bootstrapping from prior design efforts to increase efficiency and make an integrative design of architectures, optimizers, hyperparameter adaptation policies, and pretraining tasks feasible in practice.
These advances will allow the next generation of deep learning pipelines to achieve higher accuracy, lower training time, and improved ease-of-use (democratization of deeplearning). They will also allow a customization to particular design contexts, including additional objectives next to accuracy (such as robustness, memory requirements, energy consumption, latency, interpretability, training cost, uncertainty estimation, and algorithmic fairness) in order to facilitate trustworthy AI. In order to demonstrate the effectiveness of these methods, we plan to develop:
5. New state-of-the-art customized deep learning pipelines for various applications, including EEG decoding, RNA folding, and improving the reinforcement learning pipeline and deep learning on tabular data.
Wissenschaftliches Gebiet
Programm/Programme
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Thema/Themen
Finanzierungsplan
HORIZON-AG - HORIZON Action Grant Budget-BasedGastgebende Einrichtung
79098 Freiburg
Deutschland