Learn More
In this paper, we investigate how to scale up kernel methods to take on large-scale problems, on which deep neural networks have been prevailing. To this end, we leverage existing techniques and develop new ones. These techniques include approximating kernel functions with features derived from random projections, parallel training of kernel models with 100(More)
Recurrent neural networks (RNNs), including long short-term memory (LSTM) RNNs, have produced state-of-the-art results on a variety of speech recognition tasks. However, these models are often too large in size for deployment on mobile devices with memory and latency constraints. In this work, we study mechanisms for learning compact RNNs and LSTMs via(More)
Most of solid tumor cells are hypoxic and hard to trace and measure. A new compound, 4,7-bis(4-dodecylthiophen-2-yl)-5,6-dinitrobenzo[c][1,2,5]thiadiazole (BTTD-NO2), was synthesized for labeling the hypoxic cells specially in this paper. BTTD-NO2 showed no cytotoxicity to MG63 cells by MTT method. When MG63 cells were cultured with BTTD-NO2 under hypoxic(More)
Nonnegative Matrix Factorization (NMF) based on the family of β-divergences has shown to be advantageous in several signal processing and data analysis tasks. However, how to automatically select the best divergence among the family for given data remains unknown. Here we propose a new estimation criterion to resolve the problem of selecting β. Our method(More)
  • Xujun Zheng, Wencheng Zhu, Dong Liu, Hua Ai, Yan Huang, Zhiyun Lu
  • 2014
A dual-channel naphthalimide-based chemosensor for rapid and sensitive detection of fluoride ion has been developed. Upon addition of F(-), it undergoes deprotonation reaction through H-bonding interactions, and its maximum absorption wavelength is red-shifted for 214 nm to the far-red region, together with drastically quenched fluorescence. In addition, it(More)
Our current understanding is that plant species distribution in the subtropical mountain forests of Southwest China is controlled mainly by inadequate warmth. Due to abundant annual precipitation, aridity has been less considered in this context, yet rainfall here is highly seasonal, and the magnitude of drought severity at different elevations has not been(More)
A novel asymmetrical squaraine ASQ-5 bearing indoline as an end capper exhibits a low bandgap of 1.43 eV and a broad absorption band in the Vis-NIR region of 550-850 nm in thin films, hence renders solution-processed organic solar cells with an impressive Jsc of up to 11.03 mA cm(-2) and an excellent PCE of 4.29%.
An asymmetrically substituted squaraine ASQC bearing a 9-carbazyl substituent exhibits an extremely deep HOMO energy level of -5.46 eV and a relatively low bandgap of 1.65 eV, hence renders solution-processed organic solar cells with an impressive Voc of up to 1.12 V and a PCE of 2.82%.
We study large-scale kernel methods for acoustic modeling in speech recognition and compare their performance to deep neural networks (DNNs). We perform experiments on four speech recognition datasets, including the TIMIT and Broadcast News benchmark tasks, and compare these two types of models on frame-level performance metrics (accuracy, cross-entropy),(More)
A charge-transfer-featured naphthalimide derivative with a small exchange energy but a lower lying (3)ππ* state than (3)CT state is found to contribute to triplet harvesting through a P-type rather than an E-type delayed fluorescence, and could act as a quite promising host to achieve highly efficient OLEDs.