Search
Search Funnelback University
- Refined by:
- Date: 2018
31 -
40 of
243
search results for KaKaoTalk:po03 op
where 0
match all words and 243
match some words.
Results that match 1 of 2 words
-
robust.dvi
mi.eng.cam.ac.uk/~sjy/papers/heyo04.pdf20 Feb 2018: 4.2 Log-Linear Interpolation. Log-linear interpolation has been applied to languagemodel adaptation and has been shown to be equivalentto a constrained minimum Kullback-Leibler distance op-timisation problem(Klakow, -
The Effect of Cognitive Load on a Statistical Dialogue ...
mi.eng.cam.ac.uk/~sjy/papers/gtht12.pdf20 Feb 2018: 4.4 Conversational patterns. Given that the subjects felt the change of cognitiveload when they were talking to the system and op-erating the car simulator at the same time, we -
POLICY COMMITTEE FOR ADAPTATION IN MULTI-DOMAIN SPOKEN…
mi.eng.cam.ac.uk/~sjy/papers/gmsv15.pdf20 Feb 2018: 5]. Here, we address the problem ofdecision-making. Moving from a limited domain dialogue system that op-erates on a relatively modest ontology to an open domain. ... 5. EXPERIMENTAL SET-UP. In order to examine the ability of the proposed method to -
Incremental on-line adaptation of POMDP-based dialogue managers…
mi.eng.cam.ac.uk/~sjy/papers/gktb14.pdf20 Feb 2018: the Q-function estimatewe expect during the process of learning and H is a linear op-erator that captures the reward lookahead from the Q-function(see Eq. -
POMDP-based dialogue manager adaptation to extended domains M.…
mi.eng.cam.ac.uk/~sjy/papers/gbhk13a.pdf20 Feb 2018: Thisprovides increased robustness to errors in speechunderstanding and automatic dialogue policy op-timisation via reinforcement learning (Roy et al.,2000; Zhang et al., 2001; Williams and Young,2007; Young et al., -
gasic_acltslp.dvi
mi.eng.cam.ac.uk/~sjy/papers/gayo11.pdf20 Feb 2018: actions. The policy op-timisation is performed in interaction with a simulated user which gives a reward to the systemat the end of every dialogue. -
A HIERARCHICAL ATTENTION BASED MODEL FOR OFF-TOPIC SPONTANEOUSSPOKEN…
mi.eng.cam.ac.uk/~mjfg/ALTA/publications/ASRU2017/HierarchicalAttentionBased/hierarchical-attention-based.pdf24 Jan 2018: The HATMalso contains an additional 200-dimensional BiLSTM prompt-searchencoder. The ATM was trained for 5 epochs with the Adam op-timizer [19], an exponentially decaying learning rate with an -
Ghostscript wrapper for C:\Documents and Settings\mike\My…
mi.eng.cam.ac.uk/~cipolla/publications/invitedTalk/2003-MOS-handtracking.pdf13 Mar 2018: Hand Tracking Using A Quadric Surface Model. R. Cipolla1 B. Stenger1 A. Thayananthan1 P. H. S. Torr2. 1 University of Cambridge, Department of Engineering, Trumpington Street,Cambridge, CB2 1PZ, UK. 2 Microsoft Research Ltd., 7 J J Thomson Ave, -
Large scale labelled video data augmentation for semantic…
mi.eng.cam.ac.uk/~cipolla/publications/inproceedings/2017-ICCV-label-propagation.pdf13 Mar 2018: However, in contrast to image classification and somedeep learning lead problems of computer vision, semanticsegmentation (especially for autonomous driving) still op-erates on limited size datasets which do not exceed 5000labelled -
Refining Architectures of Deep Convolutional Neural Networks Sukrit…
mi.eng.cam.ac.uk/~cipolla/publications/inproceedings/2016-CVPR-refining-CNN.pdf13 Mar 2018: Please see Fig 1 for an illustration ofthese operations. We do not consider the other plausible op-erations for architectural refinement of CNN; for instance,arbitrary connection patterns between two layers ... 2. We introduce a strategy that starts with
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
Recent searches
Your search history is empty.