Filters








49,938 Hits in 2.6 sec

Stacking With Auxiliary Features

Nazneen Fatema Rajani, Raymond J. Mooney
2017 Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence  
In this paper, we propose stacking with auxiliary features that learns to fuse additional relevant information from multiple component systems as well as input instances to improve performance.  ...  We use two types of auxiliary features -- instance features and provenance features.  ...  with provenance + instance auxiliary features 0.506 0.497 Stacking with just provenance auxiliary features 0.502 0.494 Mixtures of experts (ME) model 0.494 0.489 Stacking with just instance features  ... 
doi:10.24963/ijcai.2017/367 dblp:conf/ijcai/RajaniM17 fatcat:e742hljldjbxbjmdapmfxswnd4

Stacking With Auxiliary Features [article]

Nazneen Fatema Rajani, Raymond J. Mooney
2016 arXiv   pre-print
In this paper, we propose stacking with auxiliary features that learns to fuse relevant information from multiple systems to improve performance.  ...  Auxiliary features enable the stacker to rely on systems that not just agree on an output but also the provenance of the output.  ...  with auxiliary features for ensembling.  ... 
arXiv:1605.08764v1 fatcat:5a43zcqjfvcirh35a5xzrbq35e

Stacking with Auxiliary Features for Visual Question Answering

Nazneen Fatema Rajani, Raymond Mooney
2018 Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers)  
Stacking With Auxiliary Features (SWAF) is an intelligent ensembling technique which learns to combine the results of multiple models using features of the current problem as context.  ...  We propose four categories of auxiliary features for ensembling for VQA.  ...  For the meta-classifier, we use a L1-regularized SVM classifier for generic stacking and stacking with only question/answer types as auxiliary features.  ... 
doi:10.18653/v1/n18-1201 dblp:conf/naacl/RajaniM18 fatcat:zc57ldjrjfbg3abp2pvilo26ai

Stacking With Auxiliary Features for Entity Linking in the Medical Domain

Nazneen Fatema Rajani, Mihaela Bornea, Ken Barker
2017 BioNLP 2017  
In this paper, we describe our process for building a Stacking ensemble using additional, auxiliary features for Entity Linking in the medical domain.  ...  with Auxiliary Features.  ...  Stacking With Auxiliary Features In this section we describe our algorithm and the auxiliary features used for classification. Figure 1 shows an overview of our ensembling approach.  ... 
doi:10.18653/v1/w17-2305 dblp:conf/bionlp/RajaniBB17 fatcat:g66mbnbu7jhqlcxndbs3m36haa

Tiny SSD: A Tiny Single-shot Detection Deep Convolutional Neural Network for Real-time Embedded Object Detection [article]

Alexander Wong, Mohammad Javad Shafiee, Francis Li, Brendan Chwyl
2018 arXiv   pre-print
and a non-uniform sub-network stack of highly optimized SSD-based auxiliary convolutional feature layers designed specifically to minimize model size while maintaining object detection performance.  ...  While deep neural networks have been shown in recent years to yield very powerful techniques for tackling the challenge of object detection, one of the biggest challenges with enabling such object detection  ...  stack, and ii) a non-uniform sub-network stack of highly optimized SSD-based auxiliary convolutional feature layers, with the first sub-network stack feeding into the second subnetwork stack.  ... 
arXiv:1802.06488v1 fatcat:seijmwssazgktezwu42qss2e6e

Auxiliary Stacked Denoising Autoencoder based Collaborative Filtering Recommendation

2020 KSII Transactions on Internet and Information Systems  
Firstly, we integrate auxiliary information with rating information.  ...  In this paper, we propose a novel collaborative recommendation model based on auxiliary stacked denoising autoencoder(ASDAE), the model learns effective the preferences of users from auxiliary information  ...  The Framework of ASDAE We proposed a novel approach called auxiliary stacked denoising autoencoder(ASDAE) based collaborative filtering, which integrates auxiliary information with rating information as  ... 
doi:10.3837/tiis.2020.06.001 fatcat:afeeenau3rfh3imdgwao6ugusy

Transition-based Adversarial Network for Cross-lingual Aspect Extraction

Wenya Wang, Sinno Jialin Pan
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
To solve it, we develop a novel deep model to transfer knowledge from a source language with labeled training data to a target language without any annotations.  ...  In fine-grained opinion mining, the task of aspect extraction involves the identification of explicit product features in customer reviews.  ...  An adversarial network with a language discriminator is then applied. • CrossCRF: Linear-chain CRF [Jakob and Gurevych, 2010] with non-lexical features that are similar across domains.  ... 
doi:10.24963/ijcai.2018/622 dblp:conf/ijcai/WangP18 fatcat:36pkvmr625ehxo2brwjfgrjbqe

Soft Sensor for VFA Concentration in Anaerobic Digestion Process for Treating Kitchen Waste Based on SSAE-KELM

Yuhong Wang, Shengkun Wang
2021 IEEE Access  
Besides, a combined feature selection algorithm is presented to select auxiliary variables more accurately.  ...  Given the problems of poor feature extraction and low accuracy and efficiency of the model, a stack supervised autoencoder is proposed to realize nonlinear and deep feature extraction of process data.  ...  features as the auxiliary variables of soft sensing.  ... 
doi:10.1109/access.2021.3063231 fatcat:s6sgpmoquzbipae24ksd4hwqle

Supervised and Unsupervised Ensembling for Knowledge Base Population [article]

Nazneen Fatema Rajani, Raymond J. Mooney
2016 arXiv   pre-print
We demonstrate that our combined system along with auxiliary features outperforms the best performing system for both tasks in the 2015 competition, several ensembling baselines, as well as the state-of-the-art  ...  stacking approach to ensembling KBP systems.  ...  Auxiliary Features for Stacking Along with the confidence scores, we also include auxiliary features which provide additional context Methodology Precision Recall F1 Combined stacking and constrained  ... 
arXiv:1604.04802v1 fatcat:2epjnk5acfbyhfz56enwxjjmvq

Representation Matters: Improving Perception and Exploration for Robotics [article]

Markus Wulfmeier, Arunkumar Byravan, Tim Hertweck, Irina Higgins, Ankush Gupta, Tejas Kulkarni, Malcolm Reynolds, Denis Teplyashin, Roland Hafner, Thomas Lampe, Martin Riedmiller
2021 arXiv   pre-print
The representations are evaluated in two use-cases: as input to the agent, or as a source of auxiliary tasks.  ...  Projecting high-dimensional environment observations into lower-dimensional structured representations can considerably improve data-efficiency for reinforcement learning in domains with limited data such  ...  This effect becomes visible for the smallest model class (with 2 disentangled dimensions) for the stacking and pushing tasks, where at least 4 features are necessary to represent the positions of the 2  ... 
arXiv:2011.01758v2 fatcat:wsoz3m4e4ffbxkjf4taoiyfw6e

Multi-Task Conformer with Multi-Feature Combination for Speech Emotion Recognition

Jiyoung Seo, Bowon Lee
2022 Symmetry  
In addition, we adopted multi-task learning and multi-feature combination, which showed a remarkable performance for speech emotion recognition and time-series analysis, respectively.  ...  Along with automatic speech recognition, many researchers have been actively studying speech emotion recognition, since emotion information is as crucial as the textual information for effective interactions  ...  Finally, the MFC experiment was divided into three stacking types: a single feature with STFT (FS1); a triple-feature with STFT, Mel, and MFCCs (FS3); and the MFC of all (FS5), as shown in Figure 2 .  ... 
doi:10.3390/sym14071428 fatcat:jisw3lcplbavzj4wuspkdyvnra

Augmenting Bottleneck Features of Deep Neural Network Employing Motor State for Speech Recognition at Humanoid Robots [article]

Moa Lee, Joon Hyuk Chang
2018 arXiv   pre-print
features.  ...  In this paper, a novel speech recognition system robust to ego-noise for humanoid robots is proposed, in which on/off state of the motor is employed as auxiliary information for finding the relevant input  ...  For all the bottleneck networks, stacked MFCC and auxiliary features ((13 + 2) × 11 = 165-dim.) were used for input.  ... 
arXiv:1808.08702v1 fatcat:mjhpwxvz3vbfnmke24oyq5h3dy

LOGICAL EMBEDDED PUSH-DOWN AUTOMATA IN TREE-ADJOINING GRAMMAR PARSING

Mark Johnson
1994 Computational intelligence  
Just as in the DCG extension of context-free grammars, this approach permits nodes tn be labeled with first-order terms (rather than only atomic symbols).  ...  If this succeeds, then push 8 2 in reverse order onto the top stack, insert 8 3 as a "new" stack immediately below the top stack, and finally push B1 as a new stack on top of the top stack.  ...  from the top stack of the EPDS, and nondeterministically attempt to unify it with the head a!' of each clause a' t / ! I 1 -8 2 -83 in the program.  ... 
doi:10.1111/j.1467-8640.1994.tb00012.x fatcat:wgemrpe6ujdc5ihjlprq77c2om

A Formal Reference for SCOOP [chapter]

Benjamin Morandi, Sebastian Nanz, Bertrand Meyer
2012 Lecture Notes in Computer Science  
In our formal specification, we use abstract data types with preconditions and axioms to describe the state, and introduce a number of special run-time operations to model the runtime system with our inference  ...  We extend the ADT theory with the notion of auxiliary features. Auxiliary features are convenience features that are not essential for the definition of the ADT, but nevertheless useful.  ...  The effect of a call to push env with feature or a call to push env can be undone with a call to the auxiliary command pop env .  ... 
doi:10.1007/978-3-642-25231-0_3 fatcat:6rj7c2wo7feklbwammqvjl2pva

A comprehensive operational semantics of the SCOOP programming model [article]

Benjamin Morandi, Sebastian Nanz, Bertrand Meyer
2012 arXiv   pre-print
In our formal specification, we use abstract data types with preconditions and axioms to describe the state, and introduce a number of special run-time operations to model the runtime system with our inference  ...  This work extends the ADT theory with the notion of auxiliary features. Auxiliary features are convenience features that are not essential for the definition of the ADT, but nevertheless useful.  ...  The auxiliary command push env with feature defines a state in which a processor p receives a new environment.  ... 
arXiv:1101.1038v2 fatcat:jsuio5744bcirchfdpa4rrhaf4
« Previous Showing results 1 — 15 out of 49,938 results