Zobrazeno 1 - 10
of 69
pro vyhledávání: '"Eghbal‐Zadeh, Hamid"'
Autor:
Dinu, Marius-Constantin, Holzleitner, Markus, Beck, Maximilian, Nguyen, Hoan Duc, Huber, Andrea, Eghbal-zadeh, Hamid, Moser, Bernhard A., Pereverzyev, Sergei, Hochreiter, Sepp, Zellinger, Werner
Publikováno v:
International Conference On Learning Representations (ICLR), https://openreview.net/forum?id=M95oDwJXayG, 2023
We study the problem of choosing algorithm hyper-parameters in unsupervised domain adaptation, i.e., with labeled data in a source domain and unlabeled data in a target domain, drawn from a different input distribution. We follow the strategy to comp
Externí odkaz:
http://arxiv.org/abs/2305.01281
Autor:
Koutini, Khaled, Masoudian, Shahed, Schmid, Florian, Eghbal-zadeh, Hamid, Schlüter, Jan, Widmer, Gerhard
Publikováno v:
Proceedings of Machine Learning Research v166 (2022) 65-89
The success of supervised deep learning methods is largely due to their ability to learn relevant features from raw data. Deep Neural Networks (DNNs) trained on large-scale datasets are capable of capturing a diverse set of features, and learning a r
Externí odkaz:
http://arxiv.org/abs/2211.13956
Autor:
Steinparz, Christian, Schmied, Thomas, Paischer, Fabian, Dinu, Marius-Constantin, Patil, Vihang, Bitto-Nemling, Angela, Eghbal-zadeh, Hamid, Hochreiter, Sepp
In lifelong learning, an agent learns throughout its entire life without resets, in a constantly changing environment, as we humans do. Consequently, lifelong learning comes with a plethora of research problems such as continual domain shifts, which
Externí odkaz:
http://arxiv.org/abs/2207.05742
Autor:
Gauch, Martin, Beck, Maximilian, Adler, Thomas, Kotsur, Dmytro, Fiel, Stefan, Eghbal-zadeh, Hamid, Brandstetter, Johannes, Kofler, Johannes, Holzleitner, Markus, Zellinger, Werner, Klotz, Daniel, Hochreiter, Sepp, Lehner, Sebastian
Publikováno v:
Proceedings of The 1st Conference on Lifelong Learning Agents, PMLR 199:1043-1064 (2022)
We introduce SubGD, a novel few-shot learning method which is based on the recent finding that stochastic gradient descent updates tend to live in a low-dimensional parameter subspace. In experimental and theoretical analyses, we show that models con
Externí odkaz:
http://arxiv.org/abs/2206.03483
Autor:
Paischer, Fabian, Adler, Thomas, Patil, Vihang, Bitto-Nemling, Angela, Holzleitner, Markus, Lehner, Sebastian, Eghbal-zadeh, Hamid, Hochreiter, Sepp
In a partially observable Markov decision process (POMDP), an agent typically uses a representation of the past to approximate the underlying MDP. We propose to utilize a frozen Pretrained Language Transformer (PLT) for history representation and com
Externí odkaz:
http://arxiv.org/abs/2205.12258
Autor:
Schweighofer, Kajetan, Radler, Andreas, Dinu, Marius-Constantin, Hofmarcher, Markus, Patil, Vihang, Bitto-Nemling, Angela, Eghbal-zadeh, Hamid, Hochreiter, Sepp
The application of Reinforcement Learning (RL) in real world environments can be expensive or risky due to sub-optimal policies during training. In Offline RL, this problem is avoided since interactions with an environment are prohibited. Policies ar
Externí odkaz:
http://arxiv.org/abs/2111.04714
The great success of transformer-based models in natural language processing (NLP) has led to various attempts at adapting these architectures to other domains such as vision and audio. Recent work has shown that transformers can outperform Convoluti
Externí odkaz:
http://arxiv.org/abs/2110.05069
Convolutional Neural Networks (CNNs) have been dominating classification tasks in various domains, such as machine vision, machine listening, and natural language processing. In machine listening, while generally exhibiting very good generalization c
Externí odkaz:
http://arxiv.org/abs/2107.08933
In this paper, we study the performance of variants of well-known Convolutional Neural Network (CNN) architectures on different audio tasks. We show that tuning the Receptive Field (RF) of CNNs is crucial to their generalization. An insufficient RF l
Externí odkaz:
http://arxiv.org/abs/2105.12395
Deep Neural Networks are known to be very demanding in terms of computing and memory requirements. Due to the ever increasing use of embedded systems and mobile devices with a limited resource budget, designing low-complexity models without sacrifici
Externí odkaz:
http://arxiv.org/abs/2011.02955