text stringlengths 49 577 | label stringclasses 7
values | metadata list |
|---|---|---|
We use a [[ convex formulation ]] of the << multi-label Potts model >> with label costs and show that the asymmetric map-uniqueness criterion can be integrated into our formulation by means of convex constraints . | USED-FOR | [
3,
4,
7,
9
] |
We use a convex formulation of the multi-label Potts model with label costs and show that the [[ asymmetric map-uniqueness criterion ]] can be integrated into our << formulation >> by means of convex constraints . | PART-OF | [
17,
19,
25,
25
] |
We use a convex formulation of the multi-label Potts model with label costs and show that the asymmetric map-uniqueness criterion can be integrated into our << formulation >> by means of [[ convex constraints ]] . | USED-FOR | [
29,
30,
25,
25
] |
By using a fast [[ primal-dual algorithm ]] we are able to handle several hundred << motion segments >> . | USED-FOR | [
4,
5,
13,
14
] |
Two main classes of [[ approaches ]] have been studied to perform << monocular nonrigid 3D reconstruction >> : Template-based methods and Non-rigid Structure from Motion techniques . | USED-FOR | [
4,
4,
10,
13
] |
Two main classes of << approaches >> have been studied to perform monocular nonrigid 3D reconstruction : [[ Template-based methods ]] and Non-rigid Structure from Motion techniques . | HYPONYM-OF | [
15,
16,
4,
4
] |
Two main classes of approaches have been studied to perform monocular nonrigid 3D reconstruction : [[ Template-based methods ]] and << Non-rigid Structure from Motion techniques >> . | CONJUNCTION | [
15,
16,
18,
22
] |
Two main classes of << approaches >> have been studied to perform monocular nonrigid 3D reconstruction : Template-based methods and [[ Non-rigid Structure from Motion techniques ]] . | HYPONYM-OF | [
18,
22,
4,
4
] |
While the first [[ ones ]] have been applied to reconstruct << poorly-textured surfaces >> , they assume the availability of a 3D shape model prior to reconstruction . | USED-FOR | [
3,
3,
9,
10
] |
While the first ones have been applied to reconstruct poorly-textured surfaces , << they >> assume the availability of a [[ 3D shape model ]] prior to reconstruction . | USED-FOR | [
18,
20,
12,
12
] |
In this paper , we introduce a [[ template-free approach ]] to reconstructing a << poorly-textured , deformable surface >> . | USED-FOR | [
7,
8,
12,
15
] |
To this end , we leverage [[ surface isometry ]] and formulate << 3D reconstruction >> as the joint problem of non-rigid image registration and depth estimation . | USED-FOR | [
6,
7,
10,
11
] |
To this end , we leverage surface isometry and formulate << 3D reconstruction >> as the [[ joint problem of non-rigid image registration and depth estimation ]] . | USED-FOR | [
14,
22,
10,
11
] |
Our experiments demonstrate that our [[ approach ]] yields much more accurate 3D reconstructions than << state-of-the-art techniques >> . | COMPARE | [
5,
5,
13,
14
] |
Our experiments demonstrate that our << approach >> yields much more accurate [[ 3D reconstructions ]] than state-of-the-art techniques . | EVALUATE-FOR | [
10,
11,
5,
5
] |
Our experiments demonstrate that our approach yields much more accurate [[ 3D reconstructions ]] than << state-of-the-art techniques >> . | EVALUATE-FOR | [
10,
11,
13,
14
] |
Many << computer vision applications >> , such as [[ image classification ]] and video indexing , are usually multi-label classification problems in which an instance can be assigned to more than one category . | HYPONYM-OF | [
7,
8,
1,
3
] |
Many computer vision applications , such as [[ image classification ]] and << video indexing >> , are usually multi-label classification problems in which an instance can be assigned to more than one category . | CONJUNCTION | [
7,
8,
10,
11
] |
Many << computer vision applications >> , such as image classification and [[ video indexing ]] , are usually multi-label classification problems in which an instance can be assigned to more than one category . | HYPONYM-OF | [
10,
11,
1,
3
] |
Many << computer vision applications >> , such as image classification and video indexing , are usually [[ multi-label classification problems ]] in which an instance can be assigned to more than one category . | USED-FOR | [
15,
17,
1,
3
] |
In this paper , we present a novel << multi-label classification approach >> with [[ hypergraph regu-larization ]] that addresses the correlations among different categories . | FEATURE-OF | [
12,
13,
8,
10
] |
Then , an improved [[ SVM like learning system ]] incorporating the hypergraph regularization , called Rank-HLapSVM , is proposed to handle the << multi-label classification problems >> . | USED-FOR | [
4,
7,
21,
23
] |
Then , an improved << SVM like learning system >> incorporating the [[ hypergraph regularization ]] , called Rank-HLapSVM , is proposed to handle the multi-label classification problems . | PART-OF | [
10,
11,
4,
7
] |
Then , an improved << SVM like learning system >> incorporating the hypergraph regularization , called [[ Rank-HLapSVM ]] , is proposed to handle the multi-label classification problems . | HYPONYM-OF | [
14,
14,
4,
7
] |
We find that the corresponding << optimization problem >> can be efficiently solved by the [[ dual coordinate descent method ]] . | USED-FOR | [
13,
16,
5,
6
] |
Many promising experimental results on the [[ real datasets ]] including ImageCLEF and Me-diaMill demonstrate the effectiveness and efficiency of the proposed << algorithm >> . | EVALUATE-FOR | [
6,
7,
20,
20
] |
Many promising experimental results on the << real datasets >> including [[ ImageCLEF ]] and Me-diaMill demonstrate the effectiveness and efficiency of the proposed algorithm . | HYPONYM-OF | [
9,
9,
6,
7
] |
Many promising experimental results on the real datasets including [[ ImageCLEF ]] and << Me-diaMill >> demonstrate the effectiveness and efficiency of the proposed algorithm . | CONJUNCTION | [
9,
9,
11,
11
] |
Many promising experimental results on the << real datasets >> including ImageCLEF and [[ Me-diaMill ]] demonstrate the effectiveness and efficiency of the proposed algorithm . | HYPONYM-OF | [
11,
11,
6,
7
] |
We derive a [[ convex optimization problem ]] for the task of << segmenting sequential data >> , which explicitly treats presence of outliers . | USED-FOR | [
3,
5,
10,
12
] |
We derive a convex optimization problem for the task of [[ segmenting sequential data ]] , which explicitly treats presence of << outliers >> . | USED-FOR | [
10,
12,
19,
19
] |
We describe two [[ algorithms ]] for solving this << problem >> , one exact and one a top-down novel approach , and we derive a consistency results for the case of two segments and no outliers . | USED-FOR | [
3,
3,
7,
7
] |
<< Robustness >> to [[ outliers ]] is evaluated on two real-world tasks related to speech segmentation . | FEATURE-OF | [
2,
2,
0,
0
] |
<< Robustness >> to outliers is evaluated on two [[ real-world tasks ]] related to speech segmentation . | EVALUATE-FOR | [
7,
8,
0,
0
] |
<< Robustness >> to outliers is evaluated on two real-world tasks related to [[ speech segmentation ]] . | EVALUATE-FOR | [
11,
12,
0,
0
] |
Robustness to outliers is evaluated on two << real-world tasks >> related to [[ speech segmentation ]] . | FEATURE-OF | [
11,
12,
7,
8
] |
Our [[ algorithms ]] outperform << baseline seg-mentation algorithms >> . | COMPARE | [
1,
1,
3,
5
] |
This paper examines the properties of << feature-based partial descriptions >> built on top of [[ Halliday 's systemic networks ]] . | USED-FOR | [
13,
16,
6,
8
] |
We show that the crucial operation of [[ consistency checking ]] for such << descriptions >> is NP-complete , and therefore probably intractable , but proceed to develop algorithms which can sometimes alleviate the unpleasant consequences of this intractability . | USED-FOR | [
7,
8,
11,
11
] |
We describe [[ Yoopick ]] , a << combinatorial sports prediction market >> that implements a flexible betting language , and in turn facilitates fine-grained probabilistic estimation of outcomes . | HYPONYM-OF | [
2,
2,
5,
8
] |
We describe [[ Yoopick ]] , a combinatorial sports prediction market that implements a flexible betting language , and in turn facilitates << fine-grained probabilistic estimation of outcomes >> . | USED-FOR | [
2,
2,
20,
24
] |
We describe << Yoopick >> , a combinatorial sports prediction market that implements a [[ flexible betting language ]] , and in turn facilitates fine-grained probabilistic estimation of outcomes . | USED-FOR | [
12,
14,
2,
2
] |
The goal of this paper is to discover a set of [[ discriminative patches ]] which can serve as a fully << unsupervised mid-level visual representation >> . | USED-FOR | [
11,
12,
19,
22
] |
We pose this as an << unsupervised discriminative clustering problem >> on a huge dataset of [[ image patches ]] . | USED-FOR | [
14,
15,
5,
8
] |
We use an iterative procedure which alternates between clustering and training discriminative classifiers , while applying careful [[ cross-validation ]] at each step to prevent << overfitting >> . | USED-FOR | [
17,
17,
23,
23
] |
The paper experimentally demonstrates the effectiveness of [[ discriminative patches ]] as an << unsupervised mid-level visual representation >> , suggesting that it could be used in place of visual words for many tasks . | USED-FOR | [
7,
8,
11,
14
] |
The paper experimentally demonstrates the effectiveness of << discriminative patches >> as an unsupervised mid-level visual representation , suggesting that [[ it ]] could be used in place of visual words for many tasks . | USED-FOR | [
18,
18,
7,
8
] |
Furthermore , [[ discrim-inative patches ]] can also be used in a << supervised regime >> , such as scene classification , where they demonstrate state-of-the-art performance on the MIT Indoor-67 dataset . | USED-FOR | [
2,
3,
10,
11
] |
Furthermore , discrim-inative patches can also be used in a << supervised regime >> , such as [[ scene classification ]] , where they demonstrate state-of-the-art performance on the MIT Indoor-67 dataset . | HYPONYM-OF | [
15,
16,
10,
11
] |
Furthermore , discrim-inative patches can also be used in a supervised regime , such as scene classification , where << they >> demonstrate state-of-the-art performance on the [[ MIT Indoor-67 dataset ]] . | EVALUATE-FOR | [
25,
27,
19,
19
] |
We investigate the utility of an [[ algorithm ]] for << translation lexicon acquisition -LRB- SABLE -RRB- >> , used previously on a very large corpus to acquire general translation lexicons , when that algorithm is applied to a much smaller corpus to produce candidates for domain-specific translation lexicons . | USED-FOR | [
6,
6,
8,
13
] |
We investigate the utility of an [[ algorithm ]] for translation lexicon acquisition -LRB- SABLE -RRB- , used previously on a very large corpus to acquire << general translation lexicons >> , when that algorithm is applied to a much smaller corpus to produce candidates for domain-specific translation lexicons . | USED-FOR | [
6,
6,
24,
26
] |
We investigate the utility of an algorithm for translation lexicon acquisition -LRB- SABLE -RRB- , used previously on a very large corpus to acquire general translation lexicons , when that [[ algorithm ]] is applied to a much smaller corpus to produce candidates for << domain-specific translation lexicons >> . | USED-FOR | [
30,
30,
42,
44
] |
This paper describes a [[ computational model ]] of << word segmentation >> and presents simulation results on realistic acquisition . | USED-FOR | [
4,
5,
7,
8
] |
This paper describes a [[ computational model ]] of word segmentation and presents simulation results on << realistic acquisition >> . | USED-FOR | [
4,
5,
14,
15
] |
In particular , we explore the capacity and limitations of << statistical learning mechanisms >> that have recently gained prominence in [[ cognitive psychology ]] and linguistics . | USED-FOR | [
19,
20,
10,
12
] |
In particular , we explore the capacity and limitations of statistical learning mechanisms that have recently gained prominence in [[ cognitive psychology ]] and << linguistics >> . | CONJUNCTION | [
19,
20,
22,
22
] |
In particular , we explore the capacity and limitations of << statistical learning mechanisms >> that have recently gained prominence in cognitive psychology and [[ linguistics ]] . | USED-FOR | [
22,
22,
10,
12
] |
In the [[ model-based policy search approach ]] to << reinforcement learning -LRB- RL -RRB- >> , policies are found using a model -LRB- or `` simulator '' -RRB- of the Markov decision process . | USED-FOR | [
2,
5,
7,
11
] |
In the model-based policy search approach to reinforcement learning -LRB- RL -RRB- , << policies >> are found using a model -LRB- or `` simulator '' -RRB- of the [[ Markov decision process ]] . | USED-FOR | [
27,
29,
13,
13
] |
However , for << high-dimensional continuous-state tasks >> , it can be extremely difficult to build an accurate [[ model ]] , and thus often the algorithm returns a policy that works in simulation but not in real-life . | USED-FOR | [
16,
16,
3,
5
] |
However , for high-dimensional continuous-state tasks , it can be extremely difficult to build an accurate model , and thus often the [[ algorithm ]] returns a << policy >> that works in simulation but not in real-life . | USED-FOR | [
22,
22,
25,
25
] |
The other extreme , << model-free RL >> , tends to require infeasibly large numbers of [[ real-life trials ]] . | USED-FOR | [
14,
15,
4,
5
] |
In this paper , we present a << hybrid algorithm >> that requires only an [[ approximate model ]] , and only a small number of real-life trials . | USED-FOR | [
13,
14,
7,
8
] |
In this paper , we present a hybrid algorithm that requires only an << approximate model >> , and only a small number of [[ real-life trials ]] . | USED-FOR | [
22,
23,
13,
14
] |
The key idea is to successively `` ground '' the << policy evaluations >> using [[ real-life trials ]] , but to rely on the approximate model to suggest local changes . | USED-FOR | [
13,
14,
10,
11
] |
Empirical results also demonstrate that -- when given only a [[ crude model ]] and a small number of << real-life trials >> -- our algorithm can obtain near-optimal performance in the real system . | CONJUNCTION | [
10,
11,
17,
18
] |
Empirical results also demonstrate that -- when given only a [[ crude model ]] and a small number of real-life trials -- our << algorithm >> can obtain near-optimal performance in the real system . | USED-FOR | [
10,
11,
21,
21
] |
Empirical results also demonstrate that -- when given only a crude model and a small number of [[ real-life trials ]] -- our << algorithm >> can obtain near-optimal performance in the real system . | USED-FOR | [
17,
18,
21,
21
] |
Although every << natural language system >> needs a [[ computational lexicon ]] , each system puts different amounts and types of information into its lexicon according to its individual needs . | USED-FOR | [
7,
8,
2,
4
] |
This paper presents our experience in planning and building [[ COMPLEX ]] , a << computational lexicon >> designed to be a repository of shared lexical information for use by Natural Language Processing -LRB- NLP -RRB- systems . | HYPONYM-OF | [
9,
9,
12,
13
] |
This paper presents our experience in planning and building [[ COMPLEX ]] , a computational lexicon designed to be a repository of shared lexical information for use by << Natural Language Processing -LRB- NLP -RRB- systems >> . | USED-FOR | [
9,
9,
26,
32
] |
Sentence planning is a set of inter-related but distinct << tasks >> , one of which is [[ sentence scoping ]] , i.e. the choice of syntactic structure for elementary speech acts and the decision of how to combine them into one or more sentences . | PART-OF | [
15,
16,
9,
9
] |
Sentence planning is a set of inter-related but distinct tasks , one of which is sentence scoping , i.e. the choice of [[ syntactic structure ]] for elementary << speech acts >> and the decision of how to combine them into one or more sentences . | USED-FOR | [
22,
23,
26,
27
] |
In this paper , we present [[ SPoT ]] , a << sentence planner >> , and a new methodology for automatically training SPoT on the basis of feedback provided by human judges . | HYPONYM-OF | [
6,
6,
9,
10
] |
In this paper , we present SPoT , a sentence planner , and a new [[ methodology ]] for automatically training << SPoT >> on the basis of feedback provided by human judges . | USED-FOR | [
15,
15,
19,
19
] |
First , a very simple , [[ randomized sentence-plan-generator -LRB- SPG -RRB- ]] generates a potentially large list of possible << sentence plans >> for a given text-plan input . | USED-FOR | [
6,
10,
18,
19
] |
First , a very simple , << randomized sentence-plan-generator -LRB- SPG -RRB- >> generates a potentially large list of possible sentence plans for a given [[ text-plan input ]] . | USED-FOR | [
23,
24,
6,
10
] |
Second , the [[ sentence-plan-ranker -LRB- SPR -RRB- ]] ranks the list of output << sentence plans >> , and then selects the top-ranked plan . | USED-FOR | [
3,
6,
12,
13
] |
The << SPR >> uses [[ ranking rules ]] automatically learned from training data . | USED-FOR | [
3,
4,
1,
1
] |
We show that the trained [[ SPR ]] learns to select a << sentence plan >> whose rating on average is only 5 % worse than the top human-ranked sentence plan . | USED-FOR | [
5,
5,
10,
11
] |
We show that the trained SPR learns to select a [[ sentence plan ]] whose rating on average is only 5 % worse than the << top human-ranked sentence plan >> . | COMPARE | [
10,
11,
23,
26
] |
We discuss [[ maximum a posteriori estimation ]] of << continuous density hidden Markov models -LRB- CDHMM -RRB- >> . | USED-FOR | [
2,
5,
7,
14
] |
The classical << MLE reestimation algorithms >> , namely the [[ forward-backward algorithm ]] and the segmental k-means algorithm , are expanded and reestimation formulas are given for HMM with Gaussian mixture observation densities . | HYPONYM-OF | [
8,
9,
2,
4
] |
The classical << MLE reestimation algorithms >> , namely the forward-backward algorithm and the [[ segmental k-means algorithm ]] , are expanded and reestimation formulas are given for HMM with Gaussian mixture observation densities . | HYPONYM-OF | [
12,
14,
2,
4
] |
The classical MLE reestimation algorithms , namely the << forward-backward algorithm >> and the [[ segmental k-means algorithm ]] , are expanded and reestimation formulas are given for HMM with Gaussian mixture observation densities . | CONJUNCTION | [
12,
14,
8,
9
] |
The classical MLE reestimation algorithms , namely the forward-backward algorithm and the segmental k-means algorithm , are expanded and [[ reestimation formulas ]] are given for << HMM with Gaussian mixture observation densities >> . | USED-FOR | [
19,
20,
24,
29
] |
Because of its adaptive nature , [[ Bayesian learning ]] serves as a unified approach for the following four << speech recognition applications >> , namely parameter smoothing , speaker adaptation , speaker group modeling and corrective training . | USED-FOR | [
6,
7,
17,
19
] |
Because of its adaptive nature , Bayesian learning serves as a unified approach for the following four << speech recognition applications >> , namely [[ parameter smoothing ]] , speaker adaptation , speaker group modeling and corrective training . | HYPONYM-OF | [
22,
23,
17,
19
] |
Because of its adaptive nature , Bayesian learning serves as a unified approach for the following four speech recognition applications , namely [[ parameter smoothing ]] , << speaker adaptation >> , speaker group modeling and corrective training . | CONJUNCTION | [
22,
23,
25,
26
] |
Because of its adaptive nature , Bayesian learning serves as a unified approach for the following four << speech recognition applications >> , namely parameter smoothing , [[ speaker adaptation ]] , speaker group modeling and corrective training . | HYPONYM-OF | [
25,
26,
17,
19
] |
Because of its adaptive nature , Bayesian learning serves as a unified approach for the following four speech recognition applications , namely parameter smoothing , [[ speaker adaptation ]] , << speaker group modeling >> and corrective training . | CONJUNCTION | [
25,
26,
28,
30
] |
Because of its adaptive nature , Bayesian learning serves as a unified approach for the following four << speech recognition applications >> , namely parameter smoothing , speaker adaptation , [[ speaker group modeling ]] and corrective training . | HYPONYM-OF | [
28,
30,
17,
19
] |
Because of its adaptive nature , Bayesian learning serves as a unified approach for the following four speech recognition applications , namely parameter smoothing , speaker adaptation , [[ speaker group modeling ]] and << corrective training >> . | CONJUNCTION | [
28,
30,
32,
33
] |
Because of its adaptive nature , Bayesian learning serves as a unified approach for the following four << speech recognition applications >> , namely parameter smoothing , speaker adaptation , speaker group modeling and [[ corrective training ]] . | HYPONYM-OF | [
32,
33,
17,
19
] |
New experimental results on all four [[ applications ]] are provided to show the effectiveness of the << MAP estimation approach >> . | EVALUATE-FOR | [
6,
6,
15,
17
] |
This paper describes a characters-based Chinese collocation system and discusses the advantages of [[ it ]] over a traditional << word-based system >> . | COMPARE | [
13,
13,
17,
18
] |
Since wordbreaks are not conventionally marked in Chinese text corpora , a << character-based collocation system >> has the dual advantages of [[ avoiding pre-processing distortion ]] and directly accessing sub-lexical information . | FEATURE-OF | [
20,
22,
12,
14
] |
Since wordbreaks are not conventionally marked in Chinese text corpora , a character-based collocation system has the dual advantages of [[ avoiding pre-processing distortion ]] and directly << accessing sub-lexical information >> . | CONJUNCTION | [
20,
22,
25,
27
] |
Since wordbreaks are not conventionally marked in Chinese text corpora , a << character-based collocation system >> has the dual advantages of avoiding pre-processing distortion and directly [[ accessing sub-lexical information ]] . | FEATURE-OF | [
25,
27,
12,
14
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.