Document (#40885)

Author
Maaten, L. van den
Title
Learning a parametric embedding by preserving local structure
Source
Proceedings of the Twelfth International Conference on Artificial Intelligence & Statistics (AI-STATS), JMLR W&CP 5, 2009. S.384-391
Year
2009
Abstract
The paper presents a new unsupervised dimensionality reduction technique, called parametric t-SNE, that learns a parametric mapping between the high-dimensional data space and the low-dimensional latent space. Parametric t-SNE learns the parametric mapping in such a way that the local structure of the data is preserved as well as possible in the latent space. We evaluate the performance of parametric t-SNE in experiments on three datasets, in which we compare it to the performance of two other unsupervised parametric dimensionality reduction techniques. The results of experiments illustrate the strong performance of parametric t-SNE, in particular, in learning settings in which the dimensionality of the latent space is relatively low.
Content
Vgl. auch: https://lvdmaaten.github.io/tsne/.
Theme
Data Mining
Visualisierung
Object
tSNE

Similar documents (content)

  1. Maaten, L. van den; Hinton, G.: Visualizing data using t-SNE (2008) 0.31
    0.31418726 = sum of:
      0.31418726 = product of:
        0.78546816 = sum of:
          0.013179798 = weight(abstract_txt:compare in 807) [ClassicSimilarity], result of:
            0.013179798 = score(doc=807,freq=1.0), product of:
              0.037460987 = queryWeight, product of:
                5.6292367 = idf(docFreq=412, maxDocs=42306)
                0.0066547184 = queryNorm
              0.3518273 = fieldWeight in 807, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                5.6292367 = idf(docFreq=412, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
          0.003751124 = weight(abstract_txt:which in 807) [ClassicSimilarity], result of:
            0.003751124 = score(doc=807,freq=1.0), product of:
              0.020421658 = queryWeight, product of:
                1.0441692 = boost
                2.938938 = idf(docFreq=6085, maxDocs=42306)
                0.0066547184 = queryNorm
              0.18368362 = fieldWeight in 807, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                2.938938 = idf(docFreq=6085, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
          0.016579488 = weight(abstract_txt:illustrate in 807) [ClassicSimilarity], result of:
            0.016579488 = score(doc=807,freq=1.0), product of:
              0.043653674 = queryWeight, product of:
                1.0794954 = boost
                6.076735 = idf(docFreq=263, maxDocs=42306)
                0.0066547184 = queryNorm
              0.37979594 = fieldWeight in 807, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                6.076735 = idf(docFreq=263, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
          0.015132773 = weight(abstract_txt:data in 807) [ClassicSimilarity], result of:
            0.015132773 = score(doc=807,freq=7.0), product of:
              0.02705389 = queryWeight, product of:
                1.2018223 = boost
                3.382671 = idf(docFreq=3904, maxDocs=42306)
                0.0066547184 = queryNorm
              0.55935663 = fieldWeight in 807, product of:
                2.6457512 = tf(freq=7.0), with freq of:
                  7.0 = termFreq=7.0
                3.382671 = idf(docFreq=3904, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
          0.053802397 = weight(abstract_txt:embedding in 807) [ClassicSimilarity], result of:
            0.053802397 = score(doc=807,freq=2.0), product of:
              0.07594476 = queryWeight, product of:
                1.4238338 = boost
                8.015098 = idf(docFreq=37, maxDocs=42306)
                0.0066547184 = queryNorm
              0.7084412 = fieldWeight in 807, product of:
                1.4142135 = tf(freq=2.0), with freq of:
                  2.0 = termFreq=2.0
                8.015098 = idf(docFreq=37, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
          0.021398962 = weight(abstract_txt:structure in 807) [ClassicSimilarity], result of:
            0.021398962 = score(doc=807,freq=3.0), product of:
              0.045206904 = queryWeight, product of:
                1.5535591 = boost
                4.372676 = idf(docFreq=1450, maxDocs=42306)
                0.0066547184 = queryNorm
              0.47335604 = fieldWeight in 807, product of:
                1.7320508 = tf(freq=3.0), with freq of:
                  3.0 = termFreq=3.0
                4.372676 = idf(docFreq=1450, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
          0.028753752 = weight(abstract_txt:mapping in 807) [ClassicSimilarity], result of:
            0.028753752 = score(doc=807,freq=1.0), product of:
              0.079392545 = queryWeight, product of:
                2.0588052 = boost
                5.794751 = idf(docFreq=349, maxDocs=42306)
                0.0066547184 = queryNorm
              0.36217195 = fieldWeight in 807, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                5.794751 = idf(docFreq=349, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
          0.092008136 = weight(abstract_txt:dimensional in 807) [ClassicSimilarity], result of:
            0.092008136 = score(doc=807,freq=4.0), product of:
              0.10860443 = queryWeight, product of:
                2.4079595 = boost
                6.777487 = idf(docFreq=130, maxDocs=42306)
                0.0066547184 = queryNorm
              0.84718585 = fieldWeight in 807, product of:
                2.0 = tf(freq=4.0), with freq of:
                  4.0 = termFreq=4.0
                6.777487 = idf(docFreq=130, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
          0.02226788 = weight(abstract_txt:performance in 807) [ClassicSimilarity], result of:
            0.02226788 = score(doc=807,freq=1.0), product of:
              0.07664192 = queryWeight, product of:
                2.477446 = boost
                4.6487103 = idf(docFreq=1100, maxDocs=42306)
                0.0066547184 = queryNorm
              0.2905444 = fieldWeight in 807, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                4.6487103 = idf(docFreq=1100, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
          0.51859385 = weight(abstract_txt:parametric in 807) [ClassicSimilarity], result of:
            0.51859385 = score(doc=807,freq=1.0), product of:
              0.8667389 = queryWeight, product of:
                13.605031 = boost
                9.573242 = idf(docFreq=7, maxDocs=42306)
                0.0066547184 = queryNorm
              0.59832764 = fieldWeight in 807, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                9.573242 = idf(docFreq=7, maxDocs=42306)
                0.0625 = fieldNorm(doc=807)
        0.4 = coord(10/25)
    
  2. Bose, I.; Chen, X.: ¬A method for extension of generative topographic mapping for fuzzy clustering (2009) 0.14
    0.13790508 = sum of:
      0.13790508 = product of:
        0.49251813 = sum of:
          0.016474746 = weight(abstract_txt:compare in 531) [ClassicSimilarity], result of:
            0.016474746 = score(doc=531,freq=1.0), product of:
              0.037460987 = queryWeight, product of:
                5.6292367 = idf(docFreq=412, maxDocs=42306)
                0.0066547184 = queryNorm
              0.4397841 = fieldWeight in 531, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                5.6292367 = idf(docFreq=412, maxDocs=42306)
                0.078125 = fieldNorm(doc=531)
          0.014299126 = weight(abstract_txt:data in 531) [ClassicSimilarity], result of:
            0.014299126 = score(doc=531,freq=4.0), product of:
              0.02705389 = queryWeight, product of:
                1.2018223 = boost
                3.382671 = idf(docFreq=3904, maxDocs=42306)
                0.0066547184 = queryNorm
              0.52854234 = fieldWeight in 531, product of:
                2.0 = tf(freq=4.0), with freq of:
                  4.0 = termFreq=4.0
                3.382671 = idf(docFreq=3904, maxDocs=42306)
                0.078125 = fieldNorm(doc=531)
          0.028640334 = weight(abstract_txt:experiments in 531) [ClassicSimilarity], result of:
            0.028640334 = score(doc=531,freq=1.0), product of:
              0.068238385 = queryWeight, product of:
                1.9087093 = boost
                5.372288 = idf(docFreq=533, maxDocs=42306)
                0.0066547184 = queryNorm
              0.41971 = fieldWeight in 531, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                5.372288 = idf(docFreq=533, maxDocs=42306)
                0.078125 = fieldNorm(doc=531)
          0.03594219 = weight(abstract_txt:mapping in 531) [ClassicSimilarity], result of:
            0.03594219 = score(doc=531,freq=1.0), product of:
              0.079392545 = queryWeight, product of:
                2.0588052 = boost
                5.794751 = idf(docFreq=349, maxDocs=42306)
                0.0066547184 = queryNorm
              0.45271492 = fieldWeight in 531, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                5.794751 = idf(docFreq=349, maxDocs=42306)
                0.078125 = fieldNorm(doc=531)
          0.16727042 = weight(abstract_txt:latent in 531) [ClassicSimilarity], result of:
            0.16727042 = score(doc=531,freq=3.0), product of:
              0.1756492 = queryWeight, product of:
                3.7505426 = boost
                7.037564 = idf(docFreq=100, maxDocs=42306)
                0.0066547184 = queryNorm
              0.9522983 = fieldWeight in 531, product of:
                1.7320508 = tf(freq=3.0), with freq of:
                  3.0 = termFreq=3.0
                7.037564 = idf(docFreq=100, maxDocs=42306)
                0.078125 = fieldNorm(doc=531)
          0.05870051 = weight(abstract_txt:space in 531) [ClassicSimilarity], result of:
            0.05870051 = score(doc=531,freq=1.0), product of:
              0.13872284 = queryWeight, product of:
                3.8487027 = boost
                5.4163146 = idf(docFreq=510, maxDocs=42306)
                0.0066547184 = queryNorm
              0.4231496 = fieldWeight in 531, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                5.4163146 = idf(docFreq=510, maxDocs=42306)
                0.078125 = fieldNorm(doc=531)
          0.17119081 = weight(abstract_txt:dimensionality in 531) [ClassicSimilarity], result of:
            0.17119081 = score(doc=531,freq=1.0), product of:
              0.25727293 = queryWeight, product of:
                4.5390825 = boost
                8.51719 = idf(docFreq=22, maxDocs=42306)
                0.0066547184 = queryNorm
              0.66540545 = fieldWeight in 531, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                8.51719 = idf(docFreq=22, maxDocs=42306)
                0.078125 = fieldNorm(doc=531)
        0.28 = coord(7/25)
    
  3. Efron, M.: Eigenvalue-based model selection during Latent Semantic Indexing (2005) 0.13
    0.13483253 = sum of:
      0.13483253 = product of:
        0.48154476 = sum of:
          0.004688905 = weight(abstract_txt:which in 4686) [ClassicSimilarity], result of:
            0.004688905 = score(doc=4686,freq=1.0), product of:
              0.020421658 = queryWeight, product of:
                1.0441692 = boost
                2.938938 = idf(docFreq=6085, maxDocs=42306)
                0.0066547184 = queryNorm
              0.22960453 = fieldWeight in 4686, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                2.938938 = idf(docFreq=6085, maxDocs=42306)
                0.078125 = fieldNorm(doc=4686)
          0.007149563 = weight(abstract_txt:data in 4686) [ClassicSimilarity], result of:
            0.007149563 = score(doc=4686,freq=1.0), product of:
              0.02705389 = queryWeight, product of:
                1.2018223 = boost
                3.382671 = idf(docFreq=3904, maxDocs=42306)
                0.0066547184 = queryNorm
              0.26427117 = fieldWeight in 4686, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                3.382671 = idf(docFreq=3904, maxDocs=42306)
                0.078125 = fieldNorm(doc=4686)
          0.020652149 = weight(abstract_txt:learning in 4686) [ClassicSimilarity], result of:
            0.020652149 = score(doc=4686,freq=1.0), product of:
              0.05487235 = queryWeight, product of:
                1.7115996 = boost
                4.8174996 = idf(docFreq=929, maxDocs=42306)
                0.0066547184 = queryNorm
              0.37636715 = fieldWeight in 4686, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                4.8174996 = idf(docFreq=929, maxDocs=42306)
                0.078125 = fieldNorm(doc=4686)
          0.02783485 = weight(abstract_txt:performance in 4686) [ClassicSimilarity], result of:
            0.02783485 = score(doc=4686,freq=1.0), product of:
              0.07664192 = queryWeight, product of:
                2.477446 = boost
                4.6487103 = idf(docFreq=1100, maxDocs=42306)
                0.0066547184 = queryNorm
              0.3631805 = fieldWeight in 4686, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                4.6487103 = idf(docFreq=1100, maxDocs=42306)
                0.078125 = fieldNorm(doc=4686)
          0.082545325 = weight(abstract_txt:unsupervised in 4686) [ClassicSimilarity], result of:
            0.082545325 = score(doc=4686,freq=1.0), product of:
              0.13819905 = queryWeight, product of:
                2.7163012 = boost
                7.645351 = idf(docFreq=54, maxDocs=42306)
                0.0066547184 = queryNorm
              0.597293 = fieldWeight in 4686, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                7.645351 = idf(docFreq=54, maxDocs=42306)
                0.078125 = fieldNorm(doc=4686)
          0.09657363 = weight(abstract_txt:latent in 4686) [ClassicSimilarity], result of:
            0.09657363 = score(doc=4686,freq=1.0), product of:
              0.1756492 = queryWeight, product of:
                3.7505426 = boost
                7.037564 = idf(docFreq=100, maxDocs=42306)
                0.0066547184 = queryNorm
              0.5498097 = fieldWeight in 4686, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                7.037564 = idf(docFreq=100, maxDocs=42306)
                0.078125 = fieldNorm(doc=4686)
          0.24210036 = weight(abstract_txt:dimensionality in 4686) [ClassicSimilarity], result of:
            0.24210036 = score(doc=4686,freq=2.0), product of:
              0.25727293 = queryWeight, product of:
                4.5390825 = boost
                8.51719 = idf(docFreq=22, maxDocs=42306)
                0.0066547184 = queryNorm
              0.9410254 = fieldWeight in 4686, product of:
                1.4142135 = tf(freq=2.0), with freq of:
                  2.0 = termFreq=2.0
                8.51719 = idf(docFreq=22, maxDocs=42306)
                0.078125 = fieldNorm(doc=4686)
        0.28 = coord(7/25)
    
  4. Haley, M.R.: On the normalization and distributional adjustment of journal ranking metrics : a simple parametric approach (2017) 0.11
    0.1124602 = sum of:
      0.1124602 = product of:
        0.93716836 = sum of:
          0.023064645 = weight(abstract_txt:compare in 572) [ClassicSimilarity], result of:
            0.023064645 = score(doc=572,freq=1.0), product of:
              0.037460987 = queryWeight, product of:
                5.6292367 = idf(docFreq=412, maxDocs=42306)
                0.0066547184 = queryNorm
              0.61569774 = fieldWeight in 572, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                5.6292367 = idf(docFreq=412, maxDocs=42306)
                0.109375 = fieldNorm(doc=572)
          0.0065644667 = weight(abstract_txt:which in 572) [ClassicSimilarity], result of:
            0.0065644667 = score(doc=572,freq=1.0), product of:
              0.020421658 = queryWeight, product of:
                1.0441692 = boost
                2.938938 = idf(docFreq=6085, maxDocs=42306)
                0.0066547184 = queryNorm
              0.32144633 = fieldWeight in 572, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                2.938938 = idf(docFreq=6085, maxDocs=42306)
                0.109375 = fieldNorm(doc=572)
          0.90753925 = weight(abstract_txt:parametric in 572) [ClassicSimilarity], result of:
            0.90753925 = score(doc=572,freq=1.0), product of:
              0.8667389 = queryWeight, product of:
                13.605031 = boost
                9.573242 = idf(docFreq=7, maxDocs=42306)
                0.0066547184 = queryNorm
              1.0470734 = fieldWeight in 572, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                9.573242 = idf(docFreq=7, maxDocs=42306)
                0.109375 = fieldNorm(doc=572)
        0.12 = coord(3/25)
    
  5. Yang, P.; Gao, W.; Tan, Q.; Wong, K.-F.: ¬A link-bridged topic model for cross-domain document classification (2013) 0.11
    0.11078513 = sum of:
      0.11078513 = product of:
        0.30773646 = sum of:
          0.009906726 = weight(abstract_txt:data in 4707) [ClassicSimilarity], result of:
            0.009906726 = score(doc=4707,freq=3.0), product of:
              0.02705389 = queryWeight, product of:
                1.2018223 = boost
                3.382671 = idf(docFreq=3904, maxDocs=42306)
                0.0066547184 = queryNorm
              0.3661849 = fieldWeight in 4707, product of:
                1.7320508 = tf(freq=3.0), with freq of:
                  3.0 = termFreq=3.0
                3.382671 = idf(docFreq=3904, maxDocs=42306)
                0.0625 = fieldNorm(doc=4707)
          0.023159038 = weight(abstract_txt:datasets in 4707) [ClassicSimilarity], result of:
            0.023159038 = score(doc=4707,freq=1.0), product of:
              0.054549035 = queryWeight, product of:
                1.2067128 = boost
                6.792872 = idf(docFreq=128, maxDocs=42306)
                0.0066547184 = queryNorm
              0.4245545 = fieldWeight in 4707, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                6.792872 = idf(docFreq=128, maxDocs=42306)
                0.0625 = fieldNorm(doc=4707)
          0.038044043 = weight(abstract_txt:embedding in 4707) [ClassicSimilarity], result of:
            0.038044043 = score(doc=4707,freq=1.0), product of:
              0.07594476 = queryWeight, product of:
                1.4238338 = boost
                8.015098 = idf(docFreq=37, maxDocs=42306)
                0.0066547184 = queryNorm
              0.5009436 = fieldWeight in 4707, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                8.015098 = idf(docFreq=37, maxDocs=42306)
                0.0625 = fieldNorm(doc=4707)
          0.012354696 = weight(abstract_txt:structure in 4707) [ClassicSimilarity], result of:
            0.012354696 = score(doc=4707,freq=1.0), product of:
              0.045206904 = queryWeight, product of:
                1.5535591 = boost
                4.372676 = idf(docFreq=1450, maxDocs=42306)
                0.0066547184 = queryNorm
              0.27329224 = fieldWeight in 4707, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                4.372676 = idf(docFreq=1450, maxDocs=42306)
                0.0625 = fieldNorm(doc=4707)
          0.01652172 = weight(abstract_txt:learning in 4707) [ClassicSimilarity], result of:
            0.01652172 = score(doc=4707,freq=1.0), product of:
              0.05487235 = queryWeight, product of:
                1.7115996 = boost
                4.8174996 = idf(docFreq=929, maxDocs=42306)
                0.0066547184 = queryNorm
              0.30109373 = fieldWeight in 4707, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                4.8174996 = idf(docFreq=929, maxDocs=42306)
                0.0625 = fieldNorm(doc=4707)
          0.022912268 = weight(abstract_txt:experiments in 4707) [ClassicSimilarity], result of:
            0.022912268 = score(doc=4707,freq=1.0), product of:
              0.068238385 = queryWeight, product of:
                1.9087093 = boost
                5.372288 = idf(docFreq=533, maxDocs=42306)
                0.0066547184 = queryNorm
              0.335768 = fieldWeight in 4707, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                5.372288 = idf(docFreq=533, maxDocs=42306)
                0.0625 = fieldNorm(doc=4707)
          0.028753752 = weight(abstract_txt:mapping in 4707) [ClassicSimilarity], result of:
            0.028753752 = score(doc=4707,freq=1.0), product of:
              0.079392545 = queryWeight, product of:
                2.0588052 = boost
                5.794751 = idf(docFreq=349, maxDocs=42306)
                0.0066547184 = queryNorm
              0.36217195 = fieldWeight in 4707, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                5.794751 = idf(docFreq=349, maxDocs=42306)
                0.0625 = fieldNorm(doc=4707)
          0.02226788 = weight(abstract_txt:performance in 4707) [ClassicSimilarity], result of:
            0.02226788 = score(doc=4707,freq=1.0), product of:
              0.07664192 = queryWeight, product of:
                2.477446 = boost
                4.6487103 = idf(docFreq=1100, maxDocs=42306)
                0.0066547184 = queryNorm
              0.2905444 = fieldWeight in 4707, product of:
                1.0 = tf(freq=1.0), with freq of:
                  1.0 = termFreq=1.0
                4.6487103 = idf(docFreq=1100, maxDocs=42306)
                0.0625 = fieldNorm(doc=4707)
          0.13381635 = weight(abstract_txt:latent in 4707) [ClassicSimilarity], result of:
            0.13381635 = score(doc=4707,freq=3.0), product of:
              0.1756492 = queryWeight, product of:
                3.7505426 = boost
                7.037564 = idf(docFreq=100, maxDocs=42306)
                0.0066547184 = queryNorm
              0.7618386 = fieldWeight in 4707, product of:
                1.7320508 = tf(freq=3.0), with freq of:
                  3.0 = termFreq=3.0
                7.037564 = idf(docFreq=100, maxDocs=42306)
                0.0625 = fieldNorm(doc=4707)
        0.36 = coord(9/25)