Zobrazeno 1 - 10
of 481
pro vyhledávání: '"Franke, Joerg"'
Autor:
Grazzi, Riccardo, Siems, Julien, Franke, Jörg K. H., Zela, Arber, Hutter, Frank, Pontil, Massimiliano
Linear Recurrent Neural Networks (LRNNs) such as Mamba, RWKV, GLA, mLSTM, and DeltaNet have emerged as efficient alternatives to Transformers in large language modeling, offering linear scaling with sequence length and improved training efficiency. H
Externí odkaz:
http://arxiv.org/abs/2411.12537
Autor:
Strangmann, Tobias, Purucker, Lennart, Franke, Jörg K. H., Rapant, Ivo, Ferreira, Fabio, Hutter, Frank
As the landscape of large language models expands, efficiently finetuning for specific tasks becomes increasingly crucial. At the same time, the landscape of parameter-efficient finetuning methods rapidly expands. Consequently, practitioners face a m
Externí odkaz:
http://arxiv.org/abs/2411.01195
Autor:
Sîmpetru, Raul C., Braun, Dominik I., Simon, Arndt U., März, Michael, Cnejevici, Vlad, de Oliveira, Daniela Souza, Weber, Nico, Walter, Jonas, Franke, Jörg, Höglinger, Daniel, Prahm, Cosima, Ponfick, Matthias, Del Vecchio, Alessandro
Restoring limb motor function in individuals with spinal cord injury (SCI), stroke, or amputation remains a critical challenge, one which affects millions worldwide. Recent studies show through surface electromyography (EMG) that spared motor neurons
Externí odkaz:
http://arxiv.org/abs/2408.07817
In this paper, we present the Fast Optimizer Benchmark (FOB), a tool designed for evaluating deep learning optimizers during their development. The benchmark supports tasks from multiple domains such as computer vision, natural language processing, a
Externí odkaz:
http://arxiv.org/abs/2406.18701
Autor:
Sukthanker, Rhea Sanjay, Zela, Arber, Staffler, Benedikt, Klein, Aaron, Purucker, Lennart, Franke, Joerg K. H., Hutter, Frank
The increasing size of language models necessitates a thorough analysis across multiple dimensions to assess trade-offs among crucial hardware metrics such as latency, energy consumption, GPU memory usage, and performance. Identifying optimal model c
Externí odkaz:
http://arxiv.org/abs/2405.10299
Accurate RNA secondary structure prediction is vital for understanding cellular regulation and disease mechanisms. Deep learning (DL) methods have surpassed traditional algorithms by predicting complex features like pseudoknots and multi-interacting
Externí odkaz:
http://arxiv.org/abs/2401.05351
This work introduces a new, distributed implementation of the Ensemble Kalman Filter (EnKF) that allows for non-sequential assimilation of large datasets in high-dimensional problems. The traditional EnKF algorithm is computationally intensive and ex
Externí odkaz:
http://arxiv.org/abs/2311.12909
Regularization is a critical component in deep learning. The most commonly used approach, weight decay, applies a constant penalty coefficient uniformly across all parameters. This may be overly restrictive for some parameters, while insufficient for
Externí odkaz:
http://arxiv.org/abs/2311.09058
Many Self-Supervised Learning (SSL) methods aim for model invariance to different image augmentations known as views. To achieve this invariance, conventional approaches make use of random sampling operations within the image augmentation pipeline. W
Externí odkaz:
http://arxiv.org/abs/2310.03940
Autor:
Koehler, Gregor, Wald, Tassilo, Ulrich, Constantin, Zimmerer, David, Jaeger, Paul F., Franke, Jörg K. H., Kohl, Simon, Isensee, Fabian, Maier-Hein, Klaus H.
Despite the remarkable success of deep learning systems over the last decade, a key difference still remains between neural network and human decision-making: As humans, we cannot only form a decision on the spot, but also ponder, revisiting an initi
Externí odkaz:
http://arxiv.org/abs/2309.07513