Filters








547 Hits in 5.5 sec

Cerebro: A Data System for Optimized Deep Learning Model Selection

Supun Nakandala, Yuhao Zhang, Arun Kumar
2020 Proceedings of the VLDB Endowment  
We present Cerebro, a new data system to raise deep net model selection throughput at scale without raising resource costs and without sacrificing reproducibility or accuracy.  ...  Alas, most ML systems focus on training one model at a time, reducing throughput and raising overall resource costs; some also sacrifice reproducibility.  ...  team at VMware, Carlo Curino, Matteo Interlandi, and Julian McAuley for their feedback on this work.  ... 
dblp:journals/pvldb/NakandalaZK20 fatcat:u4fx7xei7ngh3ipwpoqqdwya4m

Cerebro: A Layered Data Platform for Scalable Deep Learning

Arun Kumar, Supun Nakandala, Yuhao Zhang, Side Li, Advitya Gemawat, Kabir Nagrecha
2021 Conference on Innovative Data Systems Research  
Deep learning (DL) is gaining popularity across many domains thanks to tools such as TensorFlow and easier access to GPUs.  ...  We elevate the DL model selection process with higherlevel APIs already inherent in practice and devise a series of novel multi-query optimization techniques to substantially raise resource efficiency.  ...  team at VMware, Joe Hellerstein, Chris Jermaine, Sam Madden, Sebastian Schelter, and Dan Suciu for their feedback on this work and/or this paper.  ... 
dblp:conf/cidr/0001NZLGN21 fatcat:vkoglfacbbhfvb6at4tvdbzdmu

Hydra: A System for Large Multi-Model Deep Learning [article]

Kabir Nagrecha, Arun Kumar
2022 arXiv   pre-print
Scaling up model depth and size is now a common approach to raise accuracy in many deep learning (DL) applications, as evidenced by the widespread success of multi-billion or even trillion parameter models  ...  In this paper, we present Hydra, a system designed to tackle such challenges by enabling out-of-the-box scaling for multi-large-model DL workloads on even commodity GPUs in a resource-efficient manner.  ...  Other optimizations for DL systems that exploit multi-task execution, e.g., systems such as Model-Batch [33] , Cerebro [24] , SystemML [4] , Krypton [31] , and ASHA [26] .  ... 
arXiv:2110.08633v7 fatcat:cnmrho73w5c2zngucrrmg3rdbe

Large-Scale Brain Systems and Subcortical Relationships: The Vertically Organized Brain

Leonard F. Koziol, Lauren A. Barker, Arthur W. Joyce, Skip Hrin
2014 Applied neuropsychology. Child  
The basal ganglia anticipate and guide implicitly learned behaviors on the basis of experienced reward outcomes.  ...  The cortico-basal ganglia and the cerebro-cerebellar circuitry systems are described as fundamental to cognitive and behavioral control.  ...  Cerebro-cerebellar circuits originate within the cortex; they project to the pons, then on to the cerebellar cortex, and from there, to the deep nuclei of the cerebellum.  ... 
doi:10.1080/21622965.2014.946804 pmid:25268687 fatcat:qccrdlh6rrhvnbg5ghnyvuwpdy

Translating cancer genomics into precision medicine with artificial intelligence: applications, challenges and future perspectives

Jia Xu, Pengwei Yang, Shang Xue, Bhuvan Sharma, Marta Sanchez-Martin, Fang Wang, Kirk A. Beaty, Elinor Dehan, Baiju Parikh
2019 Human Genetics  
Integration of artificial intelligence (AI) approaches such as machine learning, deep learning, and natural language processing (NLP) to tackle the challenges of scalability and high dimensionality of  ...  In addition, the present paper highlights the challenges to AI adoption in digital healthcare with regard to data requirements, algorithmic transparency, reproducibility, and real-world assessment, and  ...  It is one term that encompasses numerous methods such as logic (rule-based), machine learning (ML), deep learning, NLP, and computer vision.  ... 
doi:10.1007/s00439-019-01970-5 fatcat:qkcyzmq4ina5jg4ieyqktjblte

Sensory Integration, Sensory Processing, and Sensory Modulation Disorders: Putative Functional Neuroanatomic Underpinnings

Leonard F. Koziol, Deborah Ely Budding, Dana Chidekel
2011 Cerebellum  
We then examine the symptoms of SID/SPD/SMD within this interactive model and in relation to their impact upon the development of inhibitory control, working memory, academic skill development, and behavioral  ...  Next, we review a dual-tiered, integrated model of brain function in order to establish neuroanatomic underpinnings with which to conceptualize the symptom presentations.  ...  It reproduces and adjusts these dynamics every time the behavior is repeated, refining the model [117, 119] .  ... 
doi:10.1007/s12311-011-0288-8 pmid:21630084 fatcat:sphvy2tp2bgjlkn5jkcz2u7ozq

Brain age prediction of healthy subjects on anatomic MRI with deep learning: going beyond with an "explainable AI" mindset [article]

Paul Herent, Simon Jegou, Gilles Wainrib, Thomas Clozel
2018 bioRxiv   pre-print
Predict brain age using various machine learning and deep learning algorithms. Define Caveat against common machine learning traps.  ...  Work on interpretability consisted in (i) proceeding on basic data visualization like correlations maps between age and voxels value, and generating (ii) weights maps of simpler models, (iii) heatmap from  ...  Deep reinforcement learning is a method inspired by dopaminergic reward system in brain, an recently outperformed alphago champion (Silver et al. 2016 ).  ... 
doi:10.1101/413302 fatcat:xm46u7mjjzerrgdxvehvdz3j34

RapiD_AI: A framework for Rapidly Deployable AI for novel disease & pandemic preparedness [article]

Alexey Youssef, Tingting Zhu, Anshul Thakur, Peter Watkinson, Peter Horby, David W Eyre, David A Clifton
2022 medRxiv   pre-print
We (i) pretrain two neural network models (Deep Neural Network and TabNet) on a large Electronic Health Records dataset representative of a general in-patient population in Oxford, UK, (ii) fine-tune using  ...  data from the first weeks of the pandemic, and (iii) simulate local deployment by testing the performance of the models on a held-out test dataset of COVID-19 patients.  ...  ., performing instance-wise feature selection), resulting in better interpretability and more efficient learning.  ... 
doi:10.1101/2022.08.09.22278600 fatcat:mbpnfxul5fblfherrgqud4l2vu

Pypes: Workflows for Processing Multimodal Neuroimaging Data

Alexandre M. Savio, Michael Schutte, Manuel Graña, Igor Yakushev
2017 Frontiers in Neuroinformatics  
Pypes has been motivated by a need for efficient and reproduceable brain PET/MRI data processing methods.  ...  Easy connection to machine learning libraries such as Nilearn and scikit-learn would allow further automatization of analyses and creation of predictive models for e.g., disease detection.  ... 
doi:10.3389/fninf.2017.00025 pmid:28443013 pmcid:PMC5387693 fatcat:g37bgt5loveqxjdqy32kkeskiq

Consensus Paper: The Cerebellum's Role in Movement and Cognition

Leonard F. Koziol, Deborah Budding, Nancy Andreasen, Stefano D'Arrigo, Sara Bulgheroni, Hiroshi Imamizu, Masao Ito, Mario Manto, Cherie Marvel, Krystal Parker, Giovanni Pezzulo, Narender Ramnani (+4 others)
2013 Cerebellum  
This paper considers the cerebellum in relation to neurocognitive development, language function, working memory, executive function, and the development of cerebellar internal control models and reflects  ...  These inferences are based on the uniformity of the cerebellum's compositional infrastructure and its apparent modular organization.  ...  that the cerebellum contributes to learning, automaticity, and behavioral adaptation through the cerebro-cerebellar circuitry system as described by Dr.  ... 
doi:10.1007/s12311-013-0511-x pmid:23996631 pmcid:PMC4089997 fatcat:qtsyhhfzh5fpjnrcijkwg4gfka

Learning Riemannian metric for disease progression modeling

Samuel Gruffaz, Pierre-Emmanuel Poulet, Etienne Maheux, Bruno Jedynak, Stanley Durrleman
2021 Neural Information Processing Systems  
They provide interpretable models at the cost of modeling assumptions on the progression profiles and their variability across subjects.  ...  A significant improvement is to embed the data in a Riemannian manifold and learn patient-specific trajectories distributed around a central geodesic.  ...  Learning a diffeomorphism is a more common task than learning a metric especially in the field of shape analysis with the LDDMM algorithm [11] , [8] and even in deep learning with the invertible networks  ... 
dblp:conf/nips/GruffazPMJD21 fatcat:x7skdbnl75fnpddvgvwupzqlri

Consensus Paper: Towards a Systems-Level View of Cerebellar Function: the Interplay Between Cerebellum, Basal Ganglia, and Cortex

Daniele Caligiore, Giovanni Pezzulo, Gianluca Baldassarre, Andreea C. Bostan, Peter L. Strick, Kenji Doya, Rick C. Helmich, Michiel Dirkx, James Houk, Henrik Jörntell, Angel Lago-Rodriguez, Joseph M. Galea (+6 others)
2016 Cerebellum  
This consensus paper gathers diverse recent views on a variety of important roles played by the cerebellum within the cerebello-basal ganglia-thalamocortical system across a range of motor and cognitive  ...  The paper includes theoretical and empirical * Daniele Caligiore  ...  reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.  ... 
doi:10.1007/s12311-016-0763-3 pmid:26873754 pmcid:PMC5243918 fatcat:3crvl2fd3fdrlejqg3nrqakq6e

Model-Driven Analysis of Eyeblink Classical Conditioning Reveals the Underlying Structure of Cerebellar Plasticity and Neuronal Activity

Alberto Antonietti, Claudia Casellato, Egidio D'Angelo, Alessandra Pedrocchi
2017 IEEE Transactions on Neural Networks and Learning Systems  
The firing of Purkinje cells (PC) and Deep Cerebellar Nuclei (DCN) changed during learning under the control of synaptic plasticity, which evolved at different rates, with a faster acquisition in the cerebellar  ...  Two subsequent sessions of EBCC acquisition and extinction were recorded and Transcranial Magnetic Stimulation (TMS) was applied on the cerebellum to alter circuit function and plasticity.  ...  Finally, we did not include cerebro-cerebellar recurrent loops in the control system.  ... 
doi:10.1109/tnnls.2016.2598190 pmid:27608482 fatcat:gvcl2mzjo5ftfmoxijvtstu3ge

The role of big brain science in the development of artificial intelligence technologies

Vykhodets Roman, Shlyapnikov Viktor
2022 Zenodo  
Strategic directions common to national projects are identified: new medical technologies for diagnostics and treatment of a wide range of diseases; technologies of deep machine learning and artificial  ...  intelligence, which are considered as the most promising in the XXIs century in terms of investment attractiveness and the impact they can have on human life and society in the context of the fourth industrial  ...  above all, the brain; development of deep machine learning and artificial intelligence technologies.  ... 
doi:10.5281/zenodo.6571103 fatcat:yfayyujkqncf7pcdgboowxdmpm

From Movement to Thought: Executive Function, Embodied Cognition, and the Cerebellum

Leonard F. Koziol, Deborah Ely Budding, Dana Chidekel
2011 Cerebellum  
Many constructs are so nonspecific and over-inclusive as to be scientifically meaningless. "Executive function" is one such term in common usage.  ...  We focus on the cerebellum's critical role in these control processes.  ...  Rather, they selectively reproduce actions that have a desired causal effect [131] .  ... 
doi:10.1007/s12311-011-0321-y pmid:22068584 fatcat:zjzgssysbzhrzmt7d2rjf7hjka
« Previous Showing results 1 — 15 out of 547 results