CAREER: Computation and Approximation in Structured Learning View Homepage


Ontology type: schema:MonetaryGrant     


Grant Info

YEARS

2013-2017

FUNDING AMOUNT

419879 USD

ABSTRACT

Machine learning is transforming the way many fields make sense of data, from engineering and science to medicine and business. Machine learning has vastly improved speech recognition, machine translation, robotic navigation and many other prediction tasks. A crucial goal of machine learning is automating intelligent processing of information: this project will focus on automatically describing videos by detecting objects, people, actions and interactions between them, and parsing documents by extracting entities, events and relationships between them. All these prediction tasks require more than just true-false or multiple-choice answers, but have an exponential number of possible answers to consider. Breaking these joint predictions up into independent decisions (for example, translating each word on its own, recognizing a phoneme at a time, detecting each object separately) ignores critical correlations and leads to poor accuracy. Structured models, such as grammars and graphical models, can capture strong dependencies but at considerable computational costs. The barrier to improving accuracy in such structured prediction problems is the prohibitive cost of inference. Structured prediction problems present a fundamental trade-off between approximation error and inference error due to computational constraints as we consider models of increasing complexity. This trade-off is poorly understood but is constantly encountered in machine learning applications. The primary outcome of this project will be a framework for addressing very large scale structured prediction using a novel coarse-to-fine architecture. This architecture will enable explicit, data-driven control of the approximation/computation trade-off. It promises to drastically advance state-of-the-art accuracy in computer vision and natural language applications and greatly enhance search and organization of documents, images, and video. The PI's plan includes an active role in the machine learning community, disseminating results through tutorials, code and data and organizing workshops. More... »

URL

http://www.nsf.gov/awardsearch/showAward?AWD_ID=1338054&HistoricalAwards=false

Related SciGraph Publications

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/2217", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/2217", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/2208", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "type": "DefinedTerm"
      }
    ], 
    "amount": {
      "currency": "USD", 
      "type": "MonetaryAmount", 
      "value": "419879"
    }, 
    "description": "Machine learning is transforming the way many fields make sense of data, from engineering and science to medicine and business. Machine learning has vastly improved speech recognition, machine translation, robotic navigation and many other prediction tasks. A crucial goal of machine learning is automating intelligent processing of information: this project will focus on automatically describing videos by detecting objects, people, actions and interactions between them, and parsing documents by extracting entities, events and relationships between them. All these prediction tasks require more than just true-false or multiple-choice answers, but have an exponential number of possible answers to consider. Breaking these joint predictions up into independent decisions (for example, translating each word on its own, recognizing a phoneme at a time, detecting each object separately) ignores critical correlations and leads to poor accuracy. Structured models, such as grammars and graphical models, can capture strong dependencies but at considerable computational costs. The barrier to improving accuracy in such structured prediction problems is the prohibitive cost of inference. Structured prediction problems present a fundamental trade-off between approximation error and inference error due to computational constraints as we consider models of increasing complexity. This trade-off is poorly understood but is constantly encountered in machine learning applications. The primary outcome of this project will be a framework for addressing very large scale structured prediction using a novel coarse-to-fine architecture. This architecture will enable explicit, data-driven control of the approximation/computation trade-off. It promises to drastically advance state-of-the-art accuracy in computer vision and natural language applications and greatly enhance search and organization of documents, images, and video. The PI's plan includes an active role in the machine learning community, disseminating results through tutorials, code and data and organizing workshops.", 
    "endDate": "2017-05-31T00:00:00Z", 
    "funder": {
      "id": "https://www.grid.ac/institutes/grid.457785.c", 
      "type": "Organization"
    }, 
    "id": "sg:grant.3488782", 
    "identifier": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "3488782"
        ]
      }, 
      {
        "name": "nsf_id", 
        "type": "PropertyValue", 
        "value": [
          "1338054"
        ]
      }
    ], 
    "inLanguage": [
      "en"
    ], 
    "keywords": [
      "approximation error", 
      "independent decisions", 
      "graphical models", 
      "search", 
      "tutorial", 
      "machine learning", 
      "joint prediction", 
      "project", 
      "structured prediction", 
      "example", 
      "words", 
      "action", 
      "structured models", 
      "many fields", 
      "relationship", 
      "fundamental trade-off", 
      "crucial goal", 
      "approximation", 
      "medicine", 
      "structured learning", 
      "control", 
      "considerable computational cost", 
      "state", 
      "prediction task", 
      "code", 
      "active role", 
      "accuracy", 
      "critical correlations", 
      "interaction", 
      "fine architecture", 
      "multiple-choice answers", 
      "robotic navigation", 
      "object", 
      "business", 
      "strong dependency", 
      "inference error", 
      "computational constraints", 
      "entity", 
      "career", 
      "document", 
      "framework", 
      "speech recognition", 
      "information", 
      "way", 
      "PI plan", 
      "machine translation", 
      "intelligent processing", 
      "many other prediction tasks", 
      "workshop", 
      "data", 
      "computer vision", 
      "computation", 
      "trade-off", 
      "video", 
      "natural language applications", 
      "poor accuracy", 
      "art accuracy", 
      "architecture", 
      "primary outcome", 
      "prohibitive cost", 
      "people", 
      "engineering", 
      "sense", 
      "structured prediction problem", 
      "complexity", 
      "phonemes", 
      "time", 
      "inference", 
      "events", 
      "results", 
      "barriers", 
      "model", 
      "machine learning applications", 
      "organization", 
      "computation trade-off", 
      "such structured prediction problems", 
      "grammar", 
      "images", 
      "science", 
      "exponential number", 
      "large scale", 
      "possible answers", 
      "community", 
      "machine"
    ], 
    "name": "CAREER: Computation and Approximation in Structured Learning", 
    "recipient": [
      {
        "id": "https://www.grid.ac/institutes/grid.34477.33", 
        "type": "Organization"
      }, 
      {
        "affiliation": {
          "id": "https://www.grid.ac/institutes/grid.34477.33", 
          "name": "University of Washington", 
          "type": "Organization"
        }, 
        "familyName": "Taskar", 
        "givenName": "Ben", 
        "id": "sg:person.010574473723.66", 
        "type": "Person"
      }, 
      {
        "member": "sg:person.010574473723.66", 
        "roleName": "PI", 
        "type": "Role"
      }, 
      {
        "affiliation": {
          "id": "https://www.grid.ac/institutes/grid.34477.33", 
          "name": "University of Washington", 
          "type": "Organization"
        }, 
        "familyName": "Farhadi", 
        "givenName": "Ali", 
        "id": "sg:person.01070776222.78", 
        "type": "Person"
      }, 
      {
        "member": "sg:person.01070776222.78", 
        "roleName": "PI", 
        "type": "Role"
      }, 
      {
        "affiliation": {
          "id": "https://www.grid.ac/institutes/grid.34477.33", 
          "name": "University of Washington", 
          "type": "Organization"
        }, 
        "familyName": "Fox", 
        "givenName": "Emily", 
        "id": "sg:person.01205224622.28", 
        "type": "Person"
      }, 
      {
        "member": "sg:person.01205224622.28", 
        "roleName": "Co-PI", 
        "type": "Role"
      }
    ], 
    "sameAs": [
      "https://app.dimensions.ai/details/grant/grant.3488782"
    ], 
    "sdDataset": "grants", 
    "sdDatePublished": "2019-03-07T12:36", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com.uberresearch.data.processor/core_data/20181219_192338/projects/base/nsf_projects_6.xml.gz", 
    "startDate": "2013-01-01T00:00:00Z", 
    "type": "MonetaryGrant", 
    "url": "http://www.nsf.gov/awardsearch/showAward?AWD_ID=1338054&HistoricalAwards=false"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/grant.3488782'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/grant.3488782'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/grant.3488782'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/grant.3488782'


 

This table displays all metadata directly associated to this object as RDF triples.

149 TRIPLES      19 PREDICATES      111 URIs      100 LITERALS      7 BLANK NODES

Subject Predicate Object
1 sg:grant.3488782 schema:about anzsrc-for:2208
2 anzsrc-for:2217
3 schema:amount N69213bd757304316802be2e4f74e6a1c
4 schema:description Machine learning is transforming the way many fields make sense of data, from engineering and science to medicine and business. Machine learning has vastly improved speech recognition, machine translation, robotic navigation and many other prediction tasks. A crucial goal of machine learning is automating intelligent processing of information: this project will focus on automatically describing videos by detecting objects, people, actions and interactions between them, and parsing documents by extracting entities, events and relationships between them. All these prediction tasks require more than just true-false or multiple-choice answers, but have an exponential number of possible answers to consider. Breaking these joint predictions up into independent decisions (for example, translating each word on its own, recognizing a phoneme at a time, detecting each object separately) ignores critical correlations and leads to poor accuracy. Structured models, such as grammars and graphical models, can capture strong dependencies but at considerable computational costs. The barrier to improving accuracy in such structured prediction problems is the prohibitive cost of inference. Structured prediction problems present a fundamental trade-off between approximation error and inference error due to computational constraints as we consider models of increasing complexity. This trade-off is poorly understood but is constantly encountered in machine learning applications. The primary outcome of this project will be a framework for addressing very large scale structured prediction using a novel coarse-to-fine architecture. This architecture will enable explicit, data-driven control of the approximation/computation trade-off. It promises to drastically advance state-of-the-art accuracy in computer vision and natural language applications and greatly enhance search and organization of documents, images, and video. The PI's plan includes an active role in the machine learning community, disseminating results through tutorials, code and data and organizing workshops.
5 schema:endDate 2017-05-31T00:00:00Z
6 schema:funder https://www.grid.ac/institutes/grid.457785.c
7 schema:identifier N4255a96afc084094be78db0a98906e9a
8 N453ee5c82f954bb7a46cb49e8536d8c1
9 schema:inLanguage en
10 schema:keywords PI plan
11 accuracy
12 action
13 active role
14 approximation
15 approximation error
16 architecture
17 art accuracy
18 barriers
19 business
20 career
21 code
22 community
23 complexity
24 computation
25 computation trade-off
26 computational constraints
27 computer vision
28 considerable computational cost
29 control
30 critical correlations
31 crucial goal
32 data
33 document
34 engineering
35 entity
36 events
37 example
38 exponential number
39 fine architecture
40 framework
41 fundamental trade-off
42 grammar
43 graphical models
44 images
45 independent decisions
46 inference
47 inference error
48 information
49 intelligent processing
50 interaction
51 joint prediction
52 large scale
53 machine
54 machine learning
55 machine learning applications
56 machine translation
57 many fields
58 many other prediction tasks
59 medicine
60 model
61 multiple-choice answers
62 natural language applications
63 object
64 organization
65 people
66 phonemes
67 poor accuracy
68 possible answers
69 prediction task
70 primary outcome
71 prohibitive cost
72 project
73 relationship
74 results
75 robotic navigation
76 science
77 search
78 sense
79 speech recognition
80 state
81 strong dependency
82 structured learning
83 structured models
84 structured prediction
85 structured prediction problem
86 such structured prediction problems
87 time
88 trade-off
89 tutorial
90 video
91 way
92 words
93 workshop
94 schema:name CAREER: Computation and Approximation in Structured Learning
95 schema:recipient N06fe1d18e2b7438ab4578a0f4a6a4a89
96 Nd37f5ffda9184f91949cd6b65f78ab1e
97 Nded41ca38694467992210f76dca26aeb
98 sg:person.010574473723.66
99 sg:person.01070776222.78
100 sg:person.01205224622.28
101 https://www.grid.ac/institutes/grid.34477.33
102 schema:sameAs https://app.dimensions.ai/details/grant/grant.3488782
103 schema:sdDatePublished 2019-03-07T12:36
104 schema:sdLicense https://scigraph.springernature.com/explorer/license/
105 schema:sdPublisher N5269c07cd0bf4cc19e16b87b0771ec06
106 schema:startDate 2013-01-01T00:00:00Z
107 schema:url http://www.nsf.gov/awardsearch/showAward?AWD_ID=1338054&HistoricalAwards=false
108 sgo:license sg:explorer/license/
109 sgo:sdDataset grants
110 rdf:type schema:MonetaryGrant
111 N06fe1d18e2b7438ab4578a0f4a6a4a89 schema:member sg:person.01070776222.78
112 schema:roleName PI
113 rdf:type schema:Role
114 N4255a96afc084094be78db0a98906e9a schema:name nsf_id
115 schema:value 1338054
116 rdf:type schema:PropertyValue
117 N453ee5c82f954bb7a46cb49e8536d8c1 schema:name dimensions_id
118 schema:value 3488782
119 rdf:type schema:PropertyValue
120 N5269c07cd0bf4cc19e16b87b0771ec06 schema:name Springer Nature - SN SciGraph project
121 rdf:type schema:Organization
122 N69213bd757304316802be2e4f74e6a1c schema:currency USD
123 schema:value 419879
124 rdf:type schema:MonetaryAmount
125 Nd37f5ffda9184f91949cd6b65f78ab1e schema:member sg:person.010574473723.66
126 schema:roleName PI
127 rdf:type schema:Role
128 Nded41ca38694467992210f76dca26aeb schema:member sg:person.01205224622.28
129 schema:roleName Co-PI
130 rdf:type schema:Role
131 anzsrc-for:2208 schema:inDefinedTermSet anzsrc-for:
132 rdf:type schema:DefinedTerm
133 anzsrc-for:2217 schema:inDefinedTermSet anzsrc-for:
134 rdf:type schema:DefinedTerm
135 sg:person.010574473723.66 schema:affiliation https://www.grid.ac/institutes/grid.34477.33
136 schema:familyName Taskar
137 schema:givenName Ben
138 rdf:type schema:Person
139 sg:person.01070776222.78 schema:affiliation https://www.grid.ac/institutes/grid.34477.33
140 schema:familyName Farhadi
141 schema:givenName Ali
142 rdf:type schema:Person
143 sg:person.01205224622.28 schema:affiliation https://www.grid.ac/institutes/grid.34477.33
144 schema:familyName Fox
145 schema:givenName Emily
146 rdf:type schema:Person
147 https://www.grid.ac/institutes/grid.34477.33 schema:name University of Washington
148 rdf:type schema:Organization
149 https://www.grid.ac/institutes/grid.457785.c schema:Organization
 




Preview window. Press ESC to close (or click here)


...