Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Visually Grounded Neural Syntax Acquisition

Yuichiroh
September 28, 2019

Visually Grounded Neural Syntax Acquisition

summarization of the paper presented at ACL 2019.
at the state-of-the-art NLP study group.

Yuichiroh

September 28, 2019
Tweet

More Decks by Yuichiroh

Other Decks in Research

Transcript

  1. ֓ཁ • ߏ੒ૉϕʔεͷߏจղੳΛڭࢣͳֶ͠श • 7JTVBMTFNBOUJDFNCFEEJOHTQBDF</HJBN FUBM > ͷख๏Λར༻͢Δ • ը૾ͱͦͷΩϟϓγϣϯΛ༻͍ɺςΩετ۠ؒͷ

    DPODSFUFOFTT BCTUSBDUOFTTείΞΛఆٛ͠ɺ ۠ؒͷ݁߹ΛΨΠυ͢Δ • ςΩετ୯ମͰͷֶशΑΓޮ཰Α͘ɺ҆ఆֶͨ͠श ͕ߦ͑Δ ࠷ઌ୺/-1ษڧձ 
  2. ؔ࿈ݚڀ  -JOHVJTUJDTUSVDUVSFJOEVDUJPOGSPNUFYU • ೥୅͸΄ͱΜͲ඼ࢺ͔Βελʔτ – ݶքʹ͍ͭͯ͸࣋ڮ͞Μ͕ݴٴ • μ΢ϯετϦʔϜλεΫ͔Β EJTUBOUTVQFSWJTJPOͰؼೲ

    ͢Δ <ʹͨ͘͞Μݚڀ> – ݴޠֶతʹଥ౰ͱࢥ͑Δߏ଄ͷಋग़ʹ੒ޭͤͣ • WJBMBOHVBHFNPEFMJOH – l5IJTBQQSPBDIIBTBDIJFWFESFNBSLBCMF QFSGPSNBODFz • 1BSTJOH3FBEJOH1SFEJDU/FUXPSL<4IFOFUBM B> • 0SEFSFE/FVSPO-45.<4IFOFUBM > ਫ໦͞Μ঺հ ࠷ઌ୺/-1ษڧձ  ͕࣌ؒͳ͍ͷͰ จݙϦετ͸࿦จΛݟ͍ͯͩ͘͞ ຊݚڀͰ͸ Language Modeling Ͱ͸ͳ͘ɺ ը૾ͱͷϚονϯάͰΨΠυ͢Δͱ͍͏ߟ͑Λಋೖ
  3. ؔ࿈ݚڀ  (SPVOEFEMBOHVBHFBDRVJTJUJPO • ը૾΍ಈըͱͦͷΩϟϓγϣϯ͔Βͷؼೲ – ͍͍ͩͨ͸ WJTVBMBUUSJCVUFT΍ BDUJPOʹ͍ͭͯਓ खͷϥϕϧ΍ϧʔϧʹج͍ͮͯؼೲ͢Δ

    • 7JTVBMTFNBOUJDFNCFEEJOHTQBDF</HJBN FUBM > – ը૾ͱςΩετͷϖΞΛѻ͏ηοςΟϯάͰ ͨ͘͞Μͷݚڀ͋Γ • JNBHFDBQUJPOSFUSJFWBM JNBHFDBQUJPO HFOFSBUJPO WJTVBM RVFTUJPOBOTXFSJOH ࠷ઌ୺/-1ษڧձ  ͕࣌ؒͳ͍ͷͰ จݙϦετ͸࿦จΛݟ͍ͯͩ͘͞ ຊݚڀ͸׬શͳڭࢣͳ͠ ͜ͷΞΠσΞΛआΓΔ
  4. ख๏ͷ֓ཁ ࠷ઌ୺/-1ษڧձ  [Ngiam et al., 2011] REINFORCE [Williams, 1992]

    ResNet-101 Bottom-up binary tree parsing Φ: ℝ$%&' → ℝ)*$ ͦΕͧΕͷߏ੒ૉʹ ϕΫτϧදݱ͕Ͱ͖Δ ಉۭؒ͡Ͱֶश
  5. 1BSTJOH TUFQ ࠷ઌ୺/-1ษڧձ  The selected pair is combined to

    form a single new constituent two-layer feedforward network
  6. 5SBJOJOH 5FYUVBM4USVDUVSF3FQSFTFOUBUJPOT ࠷ઌ୺/-1ษڧձ  Head-Initial Inductive Bias ޙΖଆ͕ functional wordΛؚΉͳΒ

    ͳΔ΂͘ޙ·Ͱ͚ͬͭ͘ͳ͍Ͱ͍͍ͨ a white on the lawn cat º the where … … desk º • ୯ମͰ͚ͬͭͨ͘͘ͳ͍ • ۟΍અΛ࡞͔ͬͯΒ͚͍ͬͭͨ͘ ⋅,⋅ ͱਖ਼൓ରͷείΞɻ ͭ·Γը૾ͱؔ࿈͕ബ͍ߏ੒ૉ ޙΖଆͷந৅౓ΛଌΔ
  7. ࣮ݧ • σʔλ .4$0$0 – USBJOEFWUFTU    –

    #FOFQBS <,JUBFW  ,MFJO > Λ࢖ͬͯQBSTJOH ͨ݁͠ՌΛ (0-%ͷ໦ͩͱࢥ͏ • ' POSBOEBNMZ TBNQMFEDBQUJPOT ࠷ઌ୺/-1ษڧձ  ͑ͬ…
  8. ݁Ռ̎ ࠷ઌ୺/-1ษڧձ  The high correlation between VG-NSL and the

    concreteness scores produced by Turney et al. (2011) and Brysbaert et al. (2014) supports the argument that the linguistic concept of concreteness can be acquired in an unsupervised way Compared to PRPN trained on the full training set, VG-NSL and VG-NSL+HI reach comparable performance using only 20% of thedata. VG-NSL tends to quickly become more stable as the amount of data increases, while PRPN and ON-LSTM remain less stable.
  9. ·ͱΊ • ը૾ͱΩϟϓγϣϯͷϖΞΛ࢖ͬͯจͷ໦ߏ଄Λ׬શ ڭࢣͳ͠Ͱֶश • ݴޠϞσϧΛݩʹ͢Δख๏ʹൺ΂ͯޮ཰Α҆͘ఆͨ͠ ֶश͕Ͱ͖Δ • ը૾શମͰ͸ͳ͘ɺը૾ʹ΋෦෼ߏ଄Λ༩͑ͯ FH

    -VFUBM 8VFUBM   ΞϥΠϯϝϯτ͢Ε͹΋ͬͱ͍͍͔΋Ͷ • ಋೖͨ͠ Head-Initial Inductive Bias ͷΑ͏ͳ΋ͷΛ ࣗಈతʹखʹೖΕΔʹ͸Ͳ͏ͨ͠Β͍͍ΜͩΖ͏ ࠷ઌ୺/-1ษڧձ