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)
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)
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)
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)
We study large-scale kernel methods for acoustic modeling and compare to DNNs on performance metrics related to both acoustic modeling and recognition. Measuring perplexity and frame-level classification accuracy, kernel-based acoustic models are as effective as their DNN counterparts. However, on token-error-rates DNN models can be significantly better. We(More)
Four novel iridium(III) complexes bearing biphenyl (7a-7c) or fluorenyl (7d) modified benzothiazole cyclometallate ligands are synthesized. In comparison with the yellow parent complex, bis(2-phenylbenzothiozolato-N,C(2')) iridium(III) (acetylacetonate) [(pbt)(2)Ir(acac)] (λ(PLmax) = 557 nm, φ(PL) = 0.26), 7a-7d show 20-43 nm bathochromic shifted orange or(More)
  • 1