Filters








571,635 Hits in 3.1 sec

Efficient synthesis of feature models

Nele Andersen, Krzysztof Czarnecki, Steven She, Andrzej Wąsowski
2012 Proceedings of the 16th International Software Product Line Conference on - SPLC '12 -volume 1  
We address the problem of automatic synthesis of feature models from propositional constraints. We show that this problem is NP-hard.  ...  We design efficient techniques for synthesis of models from respectively CNF and DNF formulas, showing a 10-to 1000-fold performance improvement over known techniques for realistic benchmarks.  ...  synthesis of feature models from conjunctive normal form (CNF) formulas, least 10-times faster than previously known algorithms. • An efficient algorithm for synthesis of feature models from disjunctive  ... 
doi:10.1145/2362536.2362553 dblp:conf/splc/AndersenCSW12 fatcat:etzk75i2gravjo3mhzqgirli2q

StressedNets: Efficient Feature Representations via Stress-induced Evolutionary Synthesis of Deep Neural Networks [article]

Mohammad Javad Shafiee, Brendan Chwyl, Francis Li, Rongyan Chen, Michelle Karg, Christian Scharfenberger, Alexander Wong
2018 arXiv   pre-print
Here, we further extend upon the evolutionary synthesis strategy for achieving efficient feature extraction via the introduction of a stress-induced evolutionary synthesis framework, where stress signals  ...  and improved model fidelity at a greater efficiency.  ...  ACKNOWLEDGMENTS The authors would like to thank ADC Automotive Distance Control Systems GmbH Continental, the Canada Research Chairs program, and Natural Sciences and Engineering Research Council of Canada  ... 
arXiv:1801.05387v1 fatcat:3r53yksey5e4vjfh7qq7bpltbm

The Mating Rituals of Deep Neural Networks: Learning Compact Feature Representations Through Sexual Evolutionary Synthesis

Audrey G. Chung, Mohammad Javad Shafiee, Paul Fieguth, Alexander Wong
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
In particular, the network synthesized via sexual evolutionary synthesis for MNIST had approximately double the architectural efficiency (cluster efficiency of 34.29× and synaptic efficiency of 258.37×  ...  Rather than the current asexual synthesis of networks, we aim to produce more compact feature representations by synthesizing more diverse and generalizable offspring networks in subsequent generations  ...  synthesis at generation 13 is 14.1×, further indicating the potential of sexual evolutionary synthesis to produce more efficient and compact feature representations.  ... 
doi:10.1109/iccvw.2017.147 dblp:conf/iccvw/ChungSFW17 fatcat:xzuwwgcwxbgbvocdkwyfcu76am

The Mating Rituals of Deep Neural Networks: Learning Compact Feature Representations through Sexual Evolutionary Synthesis [article]

Audrey Chung, Mohammad Javad Shafiee, Paul Fieguth, Alexander Wong
2017 arXiv   pre-print
In particular, the network synthesized via sexual evolutionary synthesis for MNIST had approximately double the architectural efficiency (cluster efficiency of 34.29X and synaptic efficiency of 258.37X  ...  Rather than the current asexual synthesis of networks, we aim to produce more compact feature representations by synthesizing more diverse and generalizable offspring networks in subsequent generations  ...  synthesis at generation 13 is 14.12×, further indicating the potential of sexual evolutionary synthesis to produce more efficient and compact feature representations.  ... 
arXiv:1709.02043v1 fatcat:g4d2lskftjf4xbxwpl6sfpzrwq

Learning Efficient Deep Feature Representations via Transgenerational Genetic Transmission of Environmental Information During Evolutionary Synthesis of Deep Neural Networks

M. J. Shafiee, E. Barshan, F. Li, B. Chwyl, M. Karg, C. Scharfenberger, A. Wong
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
maintaining modeling accuracy and a significantly more efficient feature representation.  ...  synapses during training to favor the synthesis of more efficient deep neural networks over successive generations.  ...  -fold in synapse efficiency using the original EvoNet) while maintaining modeling accuracy, thus resulting in a significantly more efficient deep feature representation.  ... 
doi:10.1109/iccvw.2017.120 dblp:conf/iccvw/ShafieeBLCKSW17 fatcat:jfqjnwnmprcjtbqgem2nrihbmq

Up to 100× Faster Data-free Knowledge Distillation [article]

Gongfan Fang, Kanya Mo, Xinchao Wang, Jie Song, Shitao Bei, Haofei Zhang, Mingli Song
2022 arXiv   pre-print
As a result, FastDFKD achieves data synthesis within only a few steps, significantly enhancing the efficiency of data-free training.  ...  Unlike prior methods that optimize a set of data independently, we propose to learn a meta-synthesizer that seeks common features as the initialization for the fast data synthesis.  ...  The advantage of common features sharing is that we do not need to synthesize them repeatedly for each instance, which significantly improves the synthesis efficiency.  ... 
arXiv:2112.06253v2 fatcat:5lp74qpajzhw5k5vnlgm3ui5li

An Efficient Subband Linear Prediction for LPCNet-Based Neural Synthesis

Yang Cui, Xi Wang, Lei He, Frank K. Soong
2020 Interspeech 2020  
In this paper, we introduce subband linear prediction in LPCNet for producing high fidelity speech more efficiently with consideration of subband correlation.  ...  A novel subband-based autoregressive model is proposed to learn the joint distribution of the subband sequences by introducing a reasonable assumption, which keeps the dependence between subbands while  ...  Based on the source-filter model, some neural vocoders can get natural speech with smaller model and higher efficiency, such as LPCNet [8] , LP-WaveNet [9] , neural source-filter model [10] , and glottal  ... 
doi:10.21437/interspeech.2020-1463 dblp:conf/interspeech/CuiWHS20 fatcat:ifsctom3h5htpbjcw2jpgkbete

Constructing a Deep Neural Network Based Spectral Model for Statistical Speech Synthesis [chapter]

Shinji Takaki, Junichi Yamagishi
2016 Smart Innovation, Systems and Technologies  
In this paper, we propose a statistical parametric speech synthesis system that models highdimensional spectral amplitudes directly using the DNN framework to improve modelling of spectral fine structures  ...  This paper presents a technique for spectral modeling using a deep neural network (DNN) for statistical parametric speech synthesis.  ...  On the basis of the general flow for constructing the statistical parametric speech synthesis systems, a part of layers of a DNN could be efficiently pre-trained.  ... 
doi:10.1007/978-3-319-28109-4_12 fatcat:oy2a773e5begpg7bmimipkyav4

Technology Library Modelling for Information-driven Circuit Synthesis

Lech Józwiak, Szymon Bieganski
2008 2008 11th EUROMICRO Conference on Digital System Design Architectures, Methods and Tools  
It focuses on an adequate technology library modelling for information-driven circuit synthesis.  ...  This paper considers major issues and requirements of circuit synthesis for the nano CMOS technologies, and discusses our new informationdriven circuit synthesis technology that satisfies these requirements  ...  In result of the above described gates' logic and physical feature modelling, a library model is constructed that can efficiently be used in the singlestep information-driven circuit synthesis.  ... 
doi:10.1109/dsd.2008.13 dblp:conf/dsd/JozwiakB08 fatcat:lnps42vt25ac3ovnodnovhhmja

Design Automation for Binarized Neural Networks: A Quantum Leap Opportunity?

Manuele Rusci, Lukas Cavigelli, Luca Benini
2018 2018 IEEE International Symposium on Circuits and Systems (ISCAS)  
With respect to other comparable techniques for deep learning near-sensor processing, our approach features a 10× higher energy efficiency.  ...  This paper presents the hardware design and synthesis of a purely combinational BNN for ultra-low power near-sensor processing.  ...  In this case, the synthesis tool plays a major role to enable the implementability of the model.  ... 
doi:10.1109/iscas.2018.8351807 dblp:conf/iscas/RusciCB18 fatcat:fkn4mwamhbacjj73c6vopudgjm

PixelFolder: An Efficient Progressive Pixel Synthesis Network for Image Generation [article]

Jing He, Yiyi Zhou, Qi Zhang, Jun Peng, Yunhang Shen, Xiaoshuai Sun, Chao Chen, Rongrong Ji
2022 arXiv   pre-print
In addition, we introduce novel pixel folding operations to further improve model efficiency while maintaining pixel-wise prior knowledge for end-to-end regression.  ...  In this paper, we propose a progressive pixel synthesis network towards efficient image generation, coined as PixelFolder.  ...  validates the motivations of our method about efficient pixel synthesis.  ... 
arXiv:2204.00833v4 fatcat:2ll5pooacnhjhljcbk43f5g2ru

Design Automation for Binarized Neural Networks: A Quantum Leap Opportunity? [article]

Manuele Rusci, Lukas Cavigelli, Luca Benini
2017 arXiv   pre-print
With respect to other comparable techniques for deep learning near-sensor processing, our approach features a 10x higher energy efficiency.  ...  This paper presents the hardware design and synthesis of a purely combinational BNN for ultra-low power near-sensor processing.  ...  In this case, the synthesis tool plays a major role to enable the implementability of the model.  ... 
arXiv:1712.01743v1 fatcat:qyh655py5jem3osk4dpjp5mx3q

A Multi-attribute Controllable Generative Model for Histopathology Image Synthesis [chapter]

Jiarong Ye, Yuan Xue, Peter Liu, Richard Zaino, Keith C. Cheng, Xiaolei Huang
2021 Lecture Notes in Computer Science  
In this work, we leverage the efficient self-attention and contrastive learning modules and build upon state-of-the-art generative adversarial networks (GANs) to achieve an attribute-aware image synthesis  ...  comparisons with state-of-the-art models as well as different variants of our model.  ...  of our model.  ... 
doi:10.1007/978-3-030-87237-3_59 fatcat:ayq4yk32dvfsjk7qsljnqcerfq

Autoregressive Models for Statistical Parametric Speech Synthesis

M. Shannon, Heiga Zen, W. Byrne
2013 IEEE Transactions on Audio, Speech, and Language Processing  
We propose using the autoregressive hidden Markov model (HMM) for speech synthesis.  ...  We show how to do efficient parameter estimation and synthesis with the autoregressive HMM and look at some of the similarities and differences between the standard approach, the trajectory HMM and the  ...  ACKNOWLEDGMENT The authors are very grateful to the organizers of the Blizzard Challenge for providing the scripts used to conduct the subjective evaluation.  ... 
doi:10.1109/tasl.2012.2227740 fatcat:k22e4lnfibe7tg5ve3ngzacwqu

Efficient DSP algorithm development for FPGA and ASIC technologies

Shiv Balakrishnan, Chris Eddington
2007 2007 IFIP International Conference on Very Large Scale Integration  
This paper discusses the challenges and requirements of creating portable algorithmic IP for FPGAs and ASICs and illustrates how an ESL synthesis methodology using the Synplify ® DSP tool can significantly  ...  Field-Programmable Gate Arrays (FPGAs), with their multi-million equivalent gate counts and DSP-centric features can offer dramatic performance increases over standard DSP chips.  ...  Algorithm Implementation Using DSP Synthesis Given this algorithmic model we can show the benefits of automatic DSP synthesis and architectural optimizations.  ... 
doi:10.1109/vlsisoc.2007.4402492 dblp:conf/vlsi/BalakrishnanE07 fatcat:ncgleqmfqngj5axxfvx76siiui
« Previous Showing results 1 — 15 out of 571,635 results