The temporal properties of pattern adaptation of relay cells induced by repeated sinusoidal drifting grating were investigated in the dorsal lateral geniculate nucleus (dLGN) of cats. The results showed that the respo...The temporal properties of pattern adaptation of relay cells induced by repeated sinusoidal drifting grating were investigated in the dorsal lateral geniculate nucleus (dLGN) of cats. The results showed that the response amplitude declined and the response latency prolonged when relay cells were pattern-adapted in dLGN, like the similar findings in visual cortex. However, in contrast to the result in cortex, the response phase of relay cells advanced. This implies that an inhibition with relatively long latency may participate in the pattern adaptation of dLGN cells and the adaptation in dLGN may be via a mechanism different from that of visual cortex.展开更多
A comprehensive analysis on the change of the total grain production and the temporal and spatial change of three main crops production（including wheat,maize and rice）,as well as the transfer trace of the center gra...A comprehensive analysis on the change of the total grain production and the temporal and spatial change of three main crops production（including wheat,maize and rice）,as well as the transfer trace of the center gravity of grain production in China were analyzed to reveal the overall developing trend of the grain production,explore the reasons and finally propose the corresponding suggestions and strategies to cope with the situation.展开更多
In this paper,we summarize recent progresses made in deep learning based acoustic models and the motivation and insights behind the surveyed techniques.We first discuss models such as recurrent neural networks(RNNs) a...In this paper,we summarize recent progresses made in deep learning based acoustic models and the motivation and insights behind the surveyed techniques.We first discuss models such as recurrent neural networks(RNNs) and convolutional neural networks(CNNs) that can effectively exploit variablelength contextual information,and their various combination with other models.We then describe models that are optimized end-to-end and emphasize on feature representations learned jointly with the rest of the system,the connectionist temporal classification(CTC) criterion,and the attention-based sequenceto-sequence translation model.We further illustrate robustness issues in speech recognition systems,and discuss acoustic model adaptation,speech enhancement and separation,and robust training strategies.We also cover modeling techniques that lead to more efficient decoding and discuss possible future directions in acoustic model research.展开更多
基金the National NaturalScience Foundation of China (Grant Nos. 97908003 and 30070257) the Foundation of the Chinese Academy of Sciences (Grant No. 39893340-03).
文摘The temporal properties of pattern adaptation of relay cells induced by repeated sinusoidal drifting grating were investigated in the dorsal lateral geniculate nucleus (dLGN) of cats. The results showed that the response amplitude declined and the response latency prolonged when relay cells were pattern-adapted in dLGN, like the similar findings in visual cortex. However, in contrast to the result in cortex, the response phase of relay cells advanced. This implies that an inhibition with relatively long latency may participate in the pattern adaptation of dLGN cells and the adaptation in dLGN may be via a mechanism different from that of visual cortex.
基金Supported by National Scientific and Technological Supporting Project(2006BAD20B05)~~
文摘A comprehensive analysis on the change of the total grain production and the temporal and spatial change of three main crops production（including wheat,maize and rice）,as well as the transfer trace of the center gravity of grain production in China were analyzed to reveal the overall developing trend of the grain production,explore the reasons and finally propose the corresponding suggestions and strategies to cope with the situation.
文摘In this paper,we summarize recent progresses made in deep learning based acoustic models and the motivation and insights behind the surveyed techniques.We first discuss models such as recurrent neural networks(RNNs) and convolutional neural networks(CNNs) that can effectively exploit variablelength contextual information,and their various combination with other models.We then describe models that are optimized end-to-end and emphasize on feature representations learned jointly with the rest of the system,the connectionist temporal classification(CTC) criterion,and the attention-based sequenceto-sequence translation model.We further illustrate robustness issues in speech recognition systems,and discuss acoustic model adaptation,speech enhancement and separation,and robust training strategies.We also cover modeling techniques that lead to more efficient decoding and discuss possible future directions in acoustic model research.
