Filters








20 Hits in 5.5 sec

Exemplar-Based Processing for Speech Recognition: An Overview

Tara Sainath, Bhuvana Ramabhadran, David Nahamoo, Dimitri Kanevsky, Dirk Compernolle, Kris Demuynck, Jort Gemmeke, Jerome Bellegarda, Shiva Sundaram
2012 IEEE Signal Processing Magazine  
In the machine learning community, these two broad categories of modeling are referred to as eager (offline) learning and lazy (instance-based or memory-based) learning [1] .  ...  S olving real-world classification and recognition problems requires a principled way of modeling the physical phenomena generating the observed data and the uncertainty in it.  ...  The underlying computational engine may be, e.g., GMMs, neural nets (NNs) [15] , or support vector machines (SVMs) [16] .  ... 
doi:10.1109/msp.2012.2208663 fatcat:uscjurhrejgctasb6sc5t4paca

Cognitive computing systems: Algorithms and applications for networks of neurosynaptic cores

Steve K. Esser, Alexander Andreopoulos, Rathinakumar Appuswamy, Pallab Datta, Davis Barch, Arnon Amir, John Arthur, Andrew Cassidy, Myron Flickner, Paul Merolla, Shyamal Chandra, Nicola Basilico (+11 others)
2013 The 2013 International Joint Conference on Neural Networks (IJCNN)  
To this end, we have developed a set of abstractions, algorithms, and applications that are natively efficient for TrueNorth.  ...  Second, we implemented ten algorithms that include convolution networks, spectral content estimators, liquid state machines, restricted Boltzmann machines, hidden Markov models, looming detection, temporal  ...  We would like to thank David Peyton for his expert assistance revising this manuscript.  ... 
doi:10.1109/ijcnn.2013.6706746 dblp:conf/ijcnn/EsserAADBAACFMCBCZZAKWRMNSM13 fatcat:bjkz56ezerg4rcxykd7uyyftsi

Clustering of multiple-event online sound collections with the codebook approach

Lluis Surós, Xavier Favory
2019 Zenodo  
For instance, a properly organized presentation of these audio collections can improve the user experience when browsing for sounds.  ...  This series of feature vectors will be the basis on further SMC work such as similarity assessment, classification, pattern recognition or machine learning.  ...  music tagging and then transferred to other music-related classification and regression tasks.  ... 
doi:10.5281/zenodo.3475480 fatcat:zbpmafthb5ef7i2r43pehpc4lm

Robust Correlated and Individual Component Analysis

Yannis Panagakis, Mihalis A. Nicolaou, Stefanos Zafeiriou, Maja Pantic
2016 IEEE Transactions on Pattern Analysis and Machine Intelligence  
In this light, we propose a method for the Robust Correlated and Individual Component Analysis (RCICA) of two sets of data in the presence of gross, sparse errors.  ...  The generality of the proposed methods is demonstrated by applying them onto 4 applications, namely i) heterogeneous face recognition, ii) multi-modal feature fusion for human behavior analysis (i.e.,  ...  Regression was performed via a Relevance Vector Machine (RVM) [54] . Given the input-output pair (x i , y i ), RVM models the function y i = w T φ(x i )+ i , i ∼ N (0, σ 2 ).  ... 
doi:10.1109/tpami.2015.2497700 pmid:26552077 fatcat:lea2y4hduzfozexfp7n57mjnua

Urban Computing

Yu Zheng, Licia Capra, Ouri Wolfson, Hai Yang
2014 ACM Transactions on Intelligent Systems and Technology  
For instance, the aforementioned anomaly detection uses human mobility data, road networks, and social media.  ...  Zheng et al. information will be delivered to the transportation authority for dispersing traffic and diagnosing the anomaly.  ...  ., simply putting these features into a feature vector and throwing them into a classification model) does not achieve the best performance.  ... 
doi:10.1145/2629592 fatcat:no5gcshbmrdfphv6ewm6wdoewq

Message from the general chair

Benjamin C. Lee
2015 2015 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)  
We propose a joint learning model which combines pairwise classification and mention clustering with Markov logic.  ...  We propose a candidate ranking model for "this-issue" anaphora resolution that explores different "issue"-specific and general abstract-anaphora features. The model is not restricted to nominal  ...  Query classification using topic models and support vector machine Dieu-Thu Le and Raffaella Bernardi Monday 6:00pm-9:00pm -3F Lobby (ICC) This paper describes a query classification system for a specialized  ... 
doi:10.1109/ispass.2015.7095776 dblp:conf/ispass/Lee15 fatcat:ehbed6nl6barfgs6pzwcvwxria

On the Opportunities and Risks of Foundation Models [article]

Rishi Bommasani, Drew A. Hudson, Ehsan Adeli, Russ Altman, Simran Arora, Sydney von Arx, Michael S. Bernstein, Jeannette Bohg, Antoine Bosselut, Emma Brunskill, Erik Brynjolfsson, Shyamal Buch (+102 others)
2021 arXiv   pre-print
In addition, we would like to especially thank Vanessa Parli for helping to organize this effort.  ...  Acknowledgments References ACKNOWLEDGMENTS We would like to thank the following people for their valuable feedback: Mohit Bansal, Boaz Barak, Yoshua Bengio, Sam Bowman, Collin Burns, Nicholas Carlini  ...  For example, CLIP and ViLBERT [Lu et al. 2019a ] are both multimodal vision-language, but differ in the precise way they are multimodal. 56 The former encodes images and text separately into vectors  ... 
arXiv:2108.07258v2 fatcat:yktkv4diyrgzzfzqlpvaiabc2m

Semantic Mapping in Video Retrieval

Maaike H.T. de Boer
2018 SIGIR Forum  
An advantage of using ontologies is that they provide a formal framework for supporting explicit, specific and machine-processable knowledge and provide inference and reasoning to infer implicit knowledge  ...  classification.  ... 
doi:10.1145/3190580.3190606 fatcat:a7agjytxhng4na47sfv7xsoy2a

Hubness in the protein sequence universe

Roman Vinzenz Feldbauer
2020 unpublished
In addition, deep networks are used to learn protein sequence vector representations, and investigated for orthologous group inference.  ...  The free open source software package "scikit-hubness" for Python implements these methods to make hubness analysis and reduction available to machine learning researchers and practitioners.  ...  This research is supported by the Austrian Science Fund (FWF): P27703 and P31988 References Angiulli, F. (2018 . Improving visualization of high-dimensional music similarity spaces.  ... 
doi:10.25365/thesis.64427 fatcat:wbai3saw7nbwtjtywtr73bbky4

The structural acoustic properties of stiffened shells

Yu Luan
2008 Journal of the Acoustical Society of America  
International technical standard for field measurements of DL2 integrates new descriptive models for open plan office acoustics, taking into account geometric proportions, presence of screens and furniture  ...  During the 1990s acoustic criteria for injury were designated based upon temporary hearing loss.  ...  This paper presents a novel class-specific support vector machine ͑CS-SVM͒ methodology for automated, specieslevel classification of small odontocetes.  ... 
doi:10.1121/1.2932806 fatcat:zohbewf2k5h7fly2hoqzfdk42e

Acoustic GIS‐based monitoring of Atlantic cod ecosystems in coastal Newfoundland

George A. Rose
2008 Journal of the Acoustical Society of America  
International technical standard for field measurements of DL2 integrates new descriptive models for open plan office acoustics, taking into account geometric proportions, presence of screens and furniture  ...  During the 1990s acoustic criteria for injury were designated based upon temporary hearing loss.  ...  This paper presents a novel class-specific support vector machine ͑CS-SVM͒ methodology for automated, specieslevel classification of small odontocetes.  ... 
doi:10.1121/1.2932540 fatcat:v6w3j7upx5duhkunch6oai5jve

Content Recommendation Through Linked Data

Iacopo Vagliano, M. Morisio
2017
. • Algorithms to produce recommendations based on statistical information techniques applied to Linked Data such as Support Vector Machine (SVM), Latent Dirichlet Allocation (LDA), Random Indexing (RI  ...  For each user, a regression or classification model is learned from a collection of items for which ratings are available. The training set consists of item feature vectors labeled with ratings.  ... 
doi:10.6092/polito/porto/2670692 fatcat:edcc3ul5ircsvcstvz3giu6tqu

Recognizing Human Needs Using Machine Learning-Powered Psychology-Based Framework [article]

Rajwa Abdullah Alharthi, University, My, University, My
2020
We design and develop need classification and regression models, with each corresponding to one of the predefined concepts that form the theoretical layered reference model; namely: 1) Need Content Recognition  ...  For a more comprehensive and deeper analysis, the Frustrated Need Intensity Estimator (FNIE) model and the Satisfied Need [...]  ...  preferable to deep learning algorithms, as the latter are • Support Vector Machine (SVM) Support Vector Machine (SVM) is a discriminative, non-probabilistic machine learning algorithm that is considered  ... 
doi:10.20381/ruor-24444 fatcat:qijfzwdcs5hdde2t4fbp5xq23u

Ivelina Nikolova and Natalia Konstantinova Organisers of the Student Workshop

Irina Temnikova, Natalia Konstantinova, Alexandra Balahur, Chris Biemann, Kevin Cohen, Darja Fišer, Najeh Hajlaoui, Laura Hasler, Sobha Lalitha, Devi, Wolfgang Maier, Preslav Nakov (+19 others)
unpublished
We both experiment with rule-based methods and machine learning approaches.  ...  In this paper, we show how our methods developed for identifying light verb constructions can be adapted to different domains and different types of texts.  ...  Ruslan Mitkov for their support of our research. We wish to thank our colleagues Alison Carminke, Noa P. Cruz Díaz, Dr.  ... 
fatcat:uxotu5b5mbh6jmf5wo4een7pda

A Study of Accomodation of Prosodic and Temporal Features in Spoken Dialogues in View of Speech Technology Applications

Spyridon Kousidis
2010
, posture, gaze and movement offer additional possibilities for improving on multimodal human-machine interaction.  ...  prosodic baselines, (e) informing classification for emotion recognition in dialogues ), (f) informing classification of dialogue acts (Wright 1999) , and (g) improving performance of ASR by exploiting  ...  %check for NaN if (isfinite(matrix(k,columnnumber))==1) %add to weigthed sum wsum = wsum + matrix(k,columnnumber)*duration; %add to duration sum sdur = sdur + duration; end end end k = k + 1; if k > n  ... 
doi:10.21427/d7vc8s fatcat:nxyrhdhtbvh6tar6hj52bhej5m
« Previous Showing results 1 — 15 out of 20 results