2,431 Hits in 5.6 sec

Refinement of the PAC model for the component-based design and specification of television based interfaces [chapter]

Panos Markopoulos, Paul Shrubsole, John de Vet
1999 Eurographics  
PAC [5] is a popular conceptual architecture for structuring user interface software in an object oriented fashion.  ...  Componentisation of software promises to deliver cost efficiency that has not been achieved through object orientation [19] .  ...  To move from design to development, this design oriented description needs to be matched with a specification of the software components that implement the design concept.  ... 
doi:10.1007/978-3-7091-6815-8_9 fatcat:6xvwiriigfbqljowlddu75bx3u

Tighter risk certificates for neural networks [article]

María Pérez-Ortiz and Omar Rivasplata and John Shawe-Taylor and Csaba Szepesvári
2021 arXiv   pre-print
This paper presents an empirical study regarding training probabilistic neural networks using training objectives derived from PAC-Bayes bounds.  ...  We present two training objectives, used here for the first time in connection with training neural networks. These two training objectives are derived from tight PAC-Bayes bounds.  ...  We propose -and experiment with-two new PBB training objectives: one derived from the PAC-Bayes-quadratic bound of Rivasplata et al. (2019) , and one derived from the PAC-Bayes-lambda bound of Thiemann  ... 
arXiv:2007.12911v3 fatcat:efoankqx6vbwvdeh34k76mijdm

NAS-OoD: Neural Architecture Search for Out-of-Distribution Generalization [article]

Haoyue Bai, Fengwei Zhou, Lanqing Hong, Nanyang Ye, S.-H. Gary Chan, Zhenguo Li
2021 arXiv   pre-print
Specifically, a data generator is learned to synthesize OoD data by maximizing losses computed by different neural architectures, while the goal for architecture search is to find the optimal architecture  ...  In this work, we propose robust Neural Architecture Search for OoD generalization (NAS-OoD), which optimizes the architecture with respect to its performance on generated OoD data by gradient descent.  ...  NASDA [33] leverages a principle framework that uses differentiable neural architecture search to derive optimal network architecture for domain adaptation tasks.  ... 
arXiv:2109.02038v1 fatcat:p2hzi5iegrfuvcnvhidjywpf2m

MixStyle Neural Networks for Domain Generalization and Adaptation [article]

Kaiyang Zhou, Yongxin Yang, Yu Qiao, Tao Xiang
2021 arXiv   pre-print
One way to improve domain generalization is to collect diverse source data from multiple relevant domains so that a CNN model is allowed to learn more domain-invariant, and hence generalizable representations  ...  In this work, we address domain generalization with MixStyle, a plug-and-play, parameter-free module that is simply inserted to shallow CNN layers and requires no modification to training objectives.  ...  To this end, we different CNN architectures: ResNet-50 [2] and OSNet [67]. train two baseline models: 1) Baseline, which combines The latter was specifically designed for re-ID.  ... 
arXiv:2107.02053v1 fatcat:6hflai4ikbehtldxbq5xxsbvme

Spatiotemporal Dynamics of Sound Representations Reveal a Hierarchical Progression of Category Selectivity [article]

Matthew X Lowe, Yalda Mohsenzadeh, Benjamin Lahner, Ian Charest, Aude Oliva, Santani Teng
2020 bioRxiv   pre-print
Here we show that the spatiotemporal progression from acoustic to semantically dominated representations is consistent with a hierarchical processing scheme.  ...  Semantic category representation was spatially specific: vocalizations were preferentially distinguished in temporal and frontal voice-selective regions and the fusiform face area; scene and object sounds  ...  from PAC. .  ... 
doi:10.1101/2020.06.12.149120 fatcat:sotlonu5tfej7bs25gfqlkgcba

Meta-CPR: Generalize to Unseen Large Number of Agents with Communication Pattern Recognition Module [article]

Wei-Cheng Tseng, Wei Wei, Da-Cheng Juan, Min Sun
2022 arXiv   pre-print
The ablation study is also provided to reason the proposed CPR design and show such design is effective.  ...  The proposed framework employs a meta-learned Communication Pattern Recognition (CPR) module to identify communication behavior and extract information that facilitates the training process.  ...  This baseline can help us to understand the effectiveness of the CPR module clearly. Inde-pAC: We use a naive actor-critic architecture to train the policy without communication.  ... 
arXiv:2112.07222v3 fatcat:n5kxpy3rhfhyzotg3shyzdqm2i

Functionality of gray-scale display workstation hardware and software in clinical radiology

B K Stewart, T Gillespy, T A Spraggins, S J Dwyer
1994 Radiographics  
From the results of this archive query, the user may select specific stud- ies to retrieve from the network.  ...  - ware architecture and design requirements for user interface software.  ... 
doi:10.1148/radiographics.14.3.8066278 pmid:8066278 fatcat:uvvnpvihtrh7veswx2kqnpmbeq

Domain Generalization via Model-Agnostic Learning of Semantic Features [article]

Qi Dou, Daniel C. Castro, Konstantinos Kamnitsas, Ben Glocker
2019 arXiv   pre-print
Globally, we align a derived soft confusion matrix to preserve general knowledge about inter-class relationships.  ...  The effectiveness of our method is demonstrated with new state-of-the-art results on two common object recognition benchmarks.  ...  Acknowledgements This project has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant No 757173, project MIRA, ERC  ... 
arXiv:1910.13580v1 fatcat:skh6bv2qm5apnksmvq2uu775hm

A concept-based retrieval system for thoracic radiology

Ricky K. Taira, David B. Johnson, Vikas Bhushan, Monica Rivera, Clement Wong, Lujia Huang, Denise R. Aberle, Melanie Greaves, Jonathan G. Goldin
1996 Journal of digital imaging  
Current digital information systems in radiology are insufficient to accommodate the retrieval needs of academicians.  ...  We describe a prototype system that supports intelligent case retrieval based on a combined specification of patient demographics, radiologic findings, and patho-Iogic diagnoses.  ...  From the contour of the image object, numeric descriptions of form, size, and composition can be derived.  ... 
doi:10.1007/bf03168565 pmid:8814766 fatcat:ubl777venfaptijenrtdcave7u

Taxonomy of architectural style usage

Simon Giesecke
2006 Proceedings of the 2006 conference on Pattern languages of programs - PLoP '06  
We use the term "architectural styles" to refer to high-level design patterns. We identified five major usages: ad-hoc, platformoriented, customized, pre-modeling and post-documentation/analysis.  ...  A taxonomy of architectural style usage is introduced, which serves to design new (agile or heavy-weight) software development methods that employ architectural styles.  ...  There has been important work on formalizing the properties of architectural styles using specifications in the general-purpose specification language Z [1] .  ... 
doi:10.1145/1415472.1415510 fatcat:kg6ibwh5nfh4dhazrxrlrl453u

Emergence of emotional appraisal signals in reinforcement learning agents

Pedro Sequeira, Francisco S. Melo, Ana Paiva
2014 Autonomous Agents and Multi-Agent Systems  
However, when designing autonomous agents, is emotional appraisal the best complement to the perceptions?  ...  a fitness metric that directly measures the performance of the agent in the underlying task in different scenarios. 6 Typical RL scenarios assume that Z = S and O(z | s, a) = δ(z, s), where δ denotes  ...  The first author acknowledges grant SFRH/BD/38681/2007 from the FCT.  ... 
doi:10.1007/s10458-014-9262-4 fatcat:nczqjwxrvffbjf6tzkx25fgrfa

Symbol-level precoding with per-antenna power constraints for the multi-beam satellite downlink

Danilo Spano, Symeon Chatzinotas, Jens Krause, Bjorn Ottersten
2016 2016 8th Advanced Satellite Multimedia Systems Conference and the 14th Signal Processing for Space Communications Workshop (ASMS/SPSC)  
A symbol-level precoding scheme is considered, where the data information is used, along with the channel state information, in order to exploit the multi-user interference and transform it into useful  ...  The proposed precoding design optimizes the system performance at the receiver side in terms of signal-to-noise ratio, whilst guaranteeing the system fairness and allowing a control over the power transmitted  ...  In spite of the possibility of using flexible amplifiers so to handle this issue, specific communication systems cannot afford this design.  ... 
doi:10.1109/asms-spsc.2016.7601542 dblp:conf/asms-spsc/SpanoCKO16 fatcat:ih7girxd7jbd3fejszuwms3qim

Intrahemispheric cortico-cortical connections of the human auditory cortex

Leila Cammoun, Jean Philippe Thiran, Alessandra Griffa, Reto Meuli, Patric Hagmann, Stephanie Clarke
2014 Brain Structure and Function  
It revealed a three-tier architecture characterized by a cascade of connections from the primary auditory cortex to six adjacent non-primary areas and from there to the superior temporal gyrus.  ...  On the supratemporal plane the pattern of connectivity was related to the architectonically defined early-stage auditory areas.  ...  Acknowledgement This work has been supported by the Swiss National Science Foundation grants 3200030-124897 and 320030B-141177 to S. Clarke. PH was supported by Leenaards Foundation.  ... 
doi:10.1007/s00429-014-0872-z pmid:25173473 fatcat:bnydmgok5bbp5l5o326y4l5ffq

Understanding Scanner Utilization with Real-Time DICOM Metadata Extraction

Pradeeban Kathiravelu, Ashish Sharma, Puneet Sharma
2021 IEEE Access  
Thus, they resort to using data records from the Radiology Information Systems (RIS).  ...  However, data acquired from RIS are prone to human errors, rendering many derived key performance metrics inadequate and inaccurate.  ...  ACKNOWLEDGMENT IRB approval was obtained to deploy the proposed framework in a secured research cluster in the department and acquire the DICOM images in pseudo real-time from the PACS.  ... 
doi:10.1109/access.2021.3050467 fatcat:xbnaosihenan3edwx5mygqm3gi

Learning to Localize Objects Using Limited Annotation, With Applications to Thoracic Diseases

Eyal Rozenberg, Daniel Freedman, Alex A. Bronstein
2021 IEEE Access  
used architectures and tasks.  ...  The architecture used in [6] is shown in Figure 1 .  ...  Bronstein is a Fellow of the IEEE for his contribution to 3D imaging and geometry processing.  ... 
doi:10.1109/access.2021.3075555 fatcat:22q6dyzbgfbjjhae27hlbmskla
« Previous Showing results 1 — 15 out of 2,431 results