Filters








13,229 Hits in 3.8 sec

Generalisation in humans and deep neural networks [article]

Robert Geirhos, Carlos R. Medina Temme, Jonas Rauber, Heiko H. Schütt, Matthias Bethge, Felix A. Wichmann
2020 arXiv   pre-print
We compare the robustness of humans and current convolutional deep neural networks (DNNs) on object recognition under twelve different types of image degradations.  ...  Thus, changes in the noise distribution between training and testing constitutes a crucial challenge to deep learning vision systems that can be systematically addressed in a lifelong machine learning  ...  for technical support, as well as Britta Lewke for the method of creating response icons and Patricia Rubisch for help with testing human observers.  ... 
arXiv:1808.08750v3 fatcat:3lykdavdlbbdpe4me6324cver4

Brain-Inspired Deep Imitation Learning for Autonomous Driving Systems [article]

Hasan Bayarov Ahmedov, Dewei Yi, Jie Sui
2021 arXiv   pre-print
Here, we design dual Neural Circuit Policy (NCP) architectures in deep neural networks based on the asymmetry of human neural networks.  ...  In the present study, we propose a novel brain-inspired deep imitation method that builds on the evidence from human brain functions, to improve the generalisation ability of deep neural networks so that  ...  Acknowledgments This work was was supported by the University of Aberdeen Internal Funding to Pump-Prime Interdisciplinary Research and Impact under grant number SF10206-57.  ... 
arXiv:2107.14654v1 fatcat:hgirkxut7fcfvcltf5pelx3an4

Application of a brain-inspired deep imitation learning algorithm in autonomous driving

Hasan Bayarov Ahmedov, Dewei Yi, Jie Sui
2021 Software Impacts  
Here, we design dual Neural Circuit Policy (NCP) architectures in DNN based on the asymmetry of human neural networks.  ...  In the present study, we propose a novel brain-inspired deep imitation method that builds on the evidence from human brain functions, to improve the generalisation ability of DNN so that autonomous driving  ...  Acknowledgement This work was supported by the University of Aberdeen, United Kingdom Internal Funding to Pump-Prime Interdisciplinary Research and Impact under grant number SF10206-57.  ... 
doi:10.1016/j.simpa.2021.100165 fatcat:yjlk2dowwvcetcbghmxpqo4bcy

If deep learning is the answer, then what is the question? [article]

Andrew Saxe, Stephanie Nelli, Christopher Summerfield
2020 arXiv   pre-print
Many researchers are excited by the possibility that deep neural networks may offer theories of perception, cognition and action for biological brains.  ...  If so, how can neuroscientists use deep networks to model and understand biological brains?  ...  Acknowledgements This work was supported by generous funding from the European Research Council (ERC Consolidator award to C.S. and Special Grant Agreement 3 of the Human Brain Project) and a Wellcome  ... 
arXiv:2004.07580v2 fatcat:2ltmlfs4xbdhvhh7qcga7rcbq4

Adversarial Robustness of Deep Learning: Theory, Algorithms, and Applications

Wenjie Ruan, Xinping Yi, Xiaowei Huang
2021 Proceedings of the 30th ACM International Conference on Information & Knowledge Management  
This tutorial will particularly highlight state-of-the-art techniques in adversarial attacks and robustness verification of deep neural networks (DNNs).  ...  We aim to provide a comprehensive overall picture about this emerging direction and enable the community to be aware of the urgency and importance of designing robust deep learning models in safety-critical  ...  = 623 (2) Structural Test Coverage Criteria for Deep Neural Networks, in ACM Transactions on Embedded Computing Systems 2018, Citation = 196 (3) Concolic Testing for Deep Neural Networks, in ASE 2018,  ... 
doi:10.1145/3459637.3482029 fatcat:ekos2t5jmfgahpim76txpf7qxu

Deep Learning for Cognitive Neuroscience [article]

Katherine R. Storrs, Nikolaus Kriegeskorte
2019 arXiv   pre-print
In the coming years, neural networks are likely to become less reliant on learning from massive labelled datasets, and more robust and generalisable in their task performance.  ...  Neural network models can now recognise images, understand text, translate languages, and play many human games at human or superhuman levels.  ...  We must seek out ways to make cognition in DNNs more flexible and generalisable, learning more ecologically plausible, and turn the predictive power of neural network models into theoretical understanding  ... 
arXiv:1903.01458v1 fatcat:64cray7ohncmjnwh3dfz65lrmi

PerceptNet: A Human Visual System Inspired Neural Network for Estimating Perceptual Distance [article]

Alexander Hepburn, Valero Laparra, Jesús Malo, Ryan McConville, Raul Santos-Rodriguez
2019 arXiv   pre-print
We also show that including a nonlinearity inspired by the human visual system in classical deep neural networks architectures can increase their ability to judge perceptual similarity.  ...  While recent works have presented deep neural networks trained to predict human perceptual quality, very few borrow any intuitions from the human visual system.  ...  This transformation is implemented as a deep neural network. We show that this network can generalise to datasets to more distortions than are present in the training set.  ... 
arXiv:1910.12548v1 fatcat:zhtuparv2fdefok2n6eimolwue

Shortcut Learning in Deep Neural Networks [article]

Robert Geirhos, Jörn-Henrik Jacobsen, Claudio Michaelis, Richard Zemel, Wieland Brendel, Matthias Bethge, Felix A. Wichmann
2020 arXiv   pre-print
Deep learning has triggered the current rise of artificial intelligence and is the workhorse of today's machine intelligence.  ...  In this perspective we seek to distil how many of deep learning's problem can be seen as different symptoms of the same underlying problem: shortcut learning.  ...  Author contributions The project was initiated by R.G. and C.M. and led by R.G. with support from C.M. and J.J.; M.B. and W.B. reshaped the initial thrust of the perspective and together with R.Z. supervised  ... 
arXiv:2004.07780v3 fatcat:nvfbgk6mwzgzvjcblgz577ffvi

Rethinking supervised learning: insights from biological learning and from calling it by its name [article]

Alex Hernandez-Garcia
2021 arXiv   pre-print
Here, we review insights about learning and supervision in nature, revisit the notion that learning and generalisation are not possible without supervision or inductive biases and argue that we will make  ...  The renaissance of artificial neural networks was catalysed by the success of classification models, tagged by the community with the broader term supervised learning.  ...  In particular, this project has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 641805.  ... 
arXiv:2012.02526v2 fatcat:2jevsrduizfinfa7vv2d2pfpmy

AIA: Artificial intelligence for art

Robert B. Lisek
2018 EVA London 2018  
"General" means that one AI program realises number of different tasks and the same code can be used in many applications. Artificial intelligence. Recurrent neural network. Reinforcement learning.  ...  Today's AI algorithms are limited in how much previous knowledge they are able to keep through each new training phase and how much they can reuse.  ...  SIBYL is presented as performance and installation that uses recurrent neural networks, deep learning and analog sound-video synthesizers.  ... 
doi:10.14236/ewic/eva2018.5 dblp:conf/eva/Lisek18 fatcat:wcpjqcrm2zbpvdtnpt63zls2um

Neural Algorithmic Reasoning [article]

Petar Veličković, Charles Blundell
2021 arXiv   pre-print
We argue that algorithms possess fundamentally different qualities to deep learning methods, and this strongly suggests that, were deep learning methods better able to mimic algorithms, generalisation  ...  and pragmatic solutions than those proposed by human computer scientists.  ...  concern is: will it work in a new situation? In other words, from training data, will the deep learning method generalise to the new situation.  ... 
arXiv:2105.02761v1 fatcat:q6kjndyq2nd2lphcdcqu67kcli

Real-Time Forecasting of Driver-Vehicle Dynamics on 3D Roads: a Deep-Learning Framework Leveraging Bayesian Optimisation [article]

Luca Paparusso, Stefano Melzi, Francesco Braghin
2022 arXiv   pre-print
Finally, the robustness of the neural network to changes in the driver and track was investigated to set guidelines for future works.  ...  The first, a neural network predictor, is based on Long Short-Term Memory autoencoders and fuses the information on the road geometry and the past driver-vehicle system dynamics to produce context-aware  ...  Proposed deep neural network.  ... 
arXiv:2103.03825v3 fatcat:6bdv2j2h5jcibjtkyjnttevzna

The Unstoppable Rise of Computational Linguistics in Deep Learning [article]

James Henderson
2020 arXiv   pre-print
In this paper, we trace the history of neural networks applied to natural language understanding tasks, and identify key contributions which the nature of language has made to the development of neural  ...  This perspective leads to predictions of the challenges facing research in deep learning architectures for natural language understanding.  ...  Acknowledgements We would like to thank Paola Merlo, Suzanne Stevenson, Ivan Titov, members of the Idiap NLU group, and the anonymous reviewers for their comments and suggestions.  ... 
arXiv:2005.06420v3 fatcat:3ekivq27bfg7tdh26iuseiydua

Computer-assisted EEG diagnostic review for idiopathic generalized epilepsy [article]

Shannon Clarke, Philippa Karoly, Ewan Nurse, Udaya Seneviratne, Janelle Taylor, Rory Knight-Sadler, Robert Kerr, Braden Moore, Patrick Hennessy, Dulini Mendis, Claire Lim, Jake Miles (+3 others)
2019 bioRxiv   pre-print
Deep, convolutional neural networks were trained to detect epileptic discharges using a pre-existing dataset of over 6000 labelled events in a cohort of 103 patients with idiopathic generalized epilepsy  ...  This study reports on a deep learning algorithm for computer-assisted EEG review.  ...  Neural networks Network structure Deep neural networks are computational models inspired by the architecture of the brain.  ... 
doi:10.1101/682112 fatcat:ohmcvikwhfd6pa3tddogtxpwl4

Adversarial Robustness of Deep Learning: Theory, Algorithms, and Applications [article]

Wenjie Ruan and Xinping Yi and Xiaowei Huang
2021 arXiv   pre-print
This tutorial will particularly highlight state-of-the-art techniques in adversarial attacks and robustness verification of deep neural networks (DNNs).  ...  We aim to provide a comprehensive overall picture about this emerging direction and enable the community to be aware of the urgency and importance of designing robust deep learning models in safety-critical  ...  their extension to recurrent neural networks [10] .  ... 
arXiv:2108.10451v1 fatcat:whz2yz2dbbflvmunjej4qknlwi
« Previous Showing results 1 — 15 out of 13,229 results