The high performance of the state-of-the-art deep neural networks(DNNs)is acquired at the cost of huge consumption of computing resources.Quantization of networks is recently recognized as a promising solution to solv...The high performance of the state-of-the-art deep neural networks(DNNs)is acquired at the cost of huge consumption of computing resources.Quantization of networks is recently recognized as a promising solution to solve the problem and significantly reduce the resource usage.However,the previous quantization works have mostly focused on the DNN inference,and there were very few works to address on the challenges of DNN training.In this paper,we leverage dynamic fixed-point(DFP)quantization algorithm and stochastic rounding(SR)strategy to develop a fully quantized 8-bit neural networks targeting low bitwidth training.The experiments show that,in comparison to the full-precision networks,the accuracy drop of our quantized convolutional neural networks(CNNs)can be less than 2%,even when applied to deep models evaluated on Image-Net dataset.Additionally,our 8-bit GNMT translation network can achieve almost identical BLEU to full-precision network.We further implement a prototype on FPGA and the synthesis shows that the low bitwidth training scheme can reduce the resource usage significantly.展开更多
Soft computing tools in the form of combination of multiple nonlinear regression and M5′ model tree were used for estimation of overtopping rate at the vertical coastal structures. For reliable and precise estimation...Soft computing tools in the form of combination of multiple nonlinear regression and M5′ model tree were used for estimation of overtopping rate at the vertical coastal structures. For reliable and precise estimation of overtopping rate, the experimental data available in the database CLASH were used. The dimensionless overtopping rate was estimated in terms of conventional dimensionless parameters including the relative crest freeboard Rc/Hs, seabed slope tanθ, deep water wave steepness S（om）, surf similarity ξ（om） and local relative water depth ht/Hs. The accuracy of the new model was compared with other existing models and also evaluated with some field measurements. The results indicated that the model presented in this paper is more accurate than other existing models. With statistical parameters, it is shown that the accuracy of predictions in the new model is better than that of other models.展开更多
文摘The high performance of the state-of-the-art deep neural networks(DNNs)is acquired at the cost of huge consumption of computing resources.Quantization of networks is recently recognized as a promising solution to solve the problem and significantly reduce the resource usage.However,the previous quantization works have mostly focused on the DNN inference,and there were very few works to address on the challenges of DNN training.In this paper,we leverage dynamic fixed-point(DFP)quantization algorithm and stochastic rounding(SR)strategy to develop a fully quantized 8-bit neural networks targeting low bitwidth training.The experiments show that,in comparison to the full-precision networks,the accuracy drop of our quantized convolutional neural networks(CNNs)can be less than 2%,even when applied to deep models evaluated on Image-Net dataset.Additionally,our 8-bit GNMT translation network can achieve almost identical BLEU to full-precision network.We further implement a prototype on FPGA and the synthesis shows that the low bitwidth training scheme can reduce the resource usage significantly.
文摘Soft computing tools in the form of combination of multiple nonlinear regression and M5′ model tree were used for estimation of overtopping rate at the vertical coastal structures. For reliable and precise estimation of overtopping rate, the experimental data available in the database CLASH were used. The dimensionless overtopping rate was estimated in terms of conventional dimensionless parameters including the relative crest freeboard Rc/Hs, seabed slope tanθ, deep water wave steepness S（om）, surf similarity ξ（om） and local relative water depth ht/Hs. The accuracy of the new model was compared with other existing models and also evaluated with some field measurements. The results indicated that the model presented in this paper is more accurate than other existing models. With statistical parameters, it is shown that the accuracy of predictions in the new model is better than that of other models.
