DOCSLIB.ORG

Inductive and Statistical Learning of Formal Grammars

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Inductive and Statistical Learning of Formal Grammars 2002 Grammar Induction Machine Learning Inductive and statistical learning of formal grammars Goal: to give the learning ability to a machine Pierre Dupont Design programs the performance of which improves over time [email protected] Inductive learning is a particular instance of machine learning • Goal: to find a general law from examples • Subproblem of theoretical computer science, artificial intelligence or pattern recognition – Typeset by FoilTEX – Pierre Dupont 2 2002 Grammar Induction 2002 Grammar Induction Outline Grammar Induction or Grammatical Inference • Grammar induction definition Grammar induction is a particular case of inductive learning • Learning paradigms The general law is represented by a formal grammar or an equivalent machine • DFA learning from positive and negative examples The set of examples, known as positive sample, is usually made of • RPNI algorithm strings or sequences over a specific alphabet • Probabilistic DFA learning A negative sample, i.e. a set of strings not belonging to the target language, • Application to a natural language task can sometimes help the induction process • Links with Markov models • Smoothing issues Data Induction Grammar • Related problems and future work aaabbb S−>aSb ab S−> λ Pierre Dupont 1 Pierre Dupont 3 2002 Grammar Induction 2002 Grammar Induction Examples Chromosome classification • Natural language sentence • Speech Centromere • Chronological series Chromosome 2a • Successive actions of a WEB user 90 80 70 60 • Successive moves during a chess game 50 grey density 40 30 0 100 200 300 400 500 600 • A musical piece 6 4 2 • A program 0 -2 -4 grey dens. derivative -6 0 100 200 300 400 500 600 • A form characterized by a chain code position along median axis "=====CDFDCBBBBBBBA==bcdc==DGFB=bccb== ...... ==cffc=CCC==cdb==BCB==dfdcb=====" • A biological sequence (DNA, proteins, ...) String of Primitives Pierre Dupont 4 Pierre Dupont 6 2002 Grammar Induction 2002 Grammar Induction Pattern Recognition A modeling hypothesis 16 G 0 '3.4cont' Data Grammar 15 '3.8cont' Generation Induction G 14 13 12 8dC 11 10 • Find G as close as possible to G0 3 2 1 9 8 4 0 7 • The induction process does not prove the existence of G0 5 6 7 6 It is a modeling hypothesis 5 4 3 2 1 0 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 8dC: 000077766676666555545444443211000710112344543311001234454311 Pierre Dupont 5 Pierre Dupont 7 2002 Grammar Induction 2002 Grammar Induction Identification in the limit • Grammar induction definition • Learning paradigms • DFA learning from positive and negative examples G 0 Data Grammar • RPNI algorithm Generation Induction d 1 G 1 • Probabilistic DFA learning d 2 G 2 • Application to a natural language task d • Links with Markov models n G* • Smoothing issues • convergence in finite time to G∗ • Related problems and future work ∗ • G is a representation of L(G0) (exact learning) Pierre Dupont 8 Pierre Dupont 10 2002 Grammar Induction 2002 Grammar Induction Learning paradigms PAC Learning G How to characterize learning? 0 Data Grammar Generation Induction d • which concept classes can or cannot be learned? 1 G 1 d 2 G 2 • what is a good example? d n G* • is it possible to learn in polynomial time? • convergence to G∗ ∗ • G is close enough to G0 with high probability ⇒ Probably Approximately Correct learning • polynomial time complexity Pierre Dupont 9 Pierre Dupont 11 2002 Grammar Induction 2002 Grammar Induction Other learnability results Define a probability distribution D on a set of strings Σ≤n • Identification in the limit in polynomial time L(G 0 ) L(G* ) – DFAs cannot be efficiently identified in the limit – unless we can ask equivalence and membership queries to an oracle Σ∗ • PAC learning P [P (L(G∗) ⊕ L(G )) < ] > 1 − δ D 0 – DFAs are not PAC learnable (under some cryptographic limitation assumption) – unless we can ask membership queries to an oracle The same unknown distribution D is used to generate the sample and to measure the error The result must hold for any distribution D (distribution free requirement) The algorithm must return an hypothesis in polynomial time 1 1 with respect to , δ, n, |R(L)| Pierre Dupont 12 Pierre Dupont 14 2002 Grammar Induction 2002 Grammar Induction Identification in the limit: good and bad news The bad one... • PAC learning with simple examples, i.e. examples drawn according to the conditional Solomonoff-Levin distribution Theorem 1. No superfinite class of languages is identifiable in the limit from −K(x|c) positive data only Pc(x) = λc2 The good one... K(x|c) denotes the Kolmogorov complexity of x given a representation c of the concept to be learned Theorem 2. Any admissible class of languages is identifiable in the limit from positive and negative data – regular languages are PACS learnable with positive examples only – but Kolmogorov complexity is not computable! Pierre Dupont 13 Pierre Dupont 15 2002 Grammar Induction 2002 Grammar Induction Cognitive relevance of learning paradigms • Grammar induction definition A largely unsolved question • Learning paradigms • DFA learning from positive and negative examples Learning paradigms seem irrelevant to model human learning: • RPNI algorithm • Gold’s identification in the limit framework has been criticized as children seem • Probabilistic DFA learning to learn natural language without negative examples • Application to a natural language task • All learning models assume a known representation class • Links with Markov models • Smoothing issues • Some learnability results are based on enumeration • Related problems and future work Pierre Dupont 16 Pierre Dupont 18 2002 Grammar Induction 2002 Grammar Induction Regular Inference from Positive and Negative Data However learning models show that: Additional hypothesis: the underlying theory is a regular grammar or, equiva- • an oracle can help lently, a finite state automaton • some examples are useless, others are good: characteristic samples ⇔ typical examples Property 1. Any regular language has a canonical automaton A(L) which is deterministic and minimal (minimal DFA) • learning well is learning efficiently Example : L = (ba∗a)∗ a • example frequency matters b b a 0 1 2 • good examples are simple examples ⇔ cognitive economy Pierre Dupont 17 Pierre Dupont 19 2002 Grammar Induction 2002 Grammar Induction A few definitions A theorem Definition 1. A positive sample S+ is structurally complete The positive data can be represented by a prefix tree acceptor (PTA) with respect to an automaton A if, when generating S+ from A: a 3 b b a • every transition of A is used at least one a 1 4 6 8 0 b a a 2 5 7 • every final state is used as accepting state of at least one string Example : {aa, abba, baa} a b Theorem 3. If the positive sample is structurally complete with respect to a b a 0 1 2 canonical automaton A(L0) then there exists a partition π of the state set of PTA Example : {ba, baa, baba, λ} such that P T A/π = A(L0) Pierre Dupont 20 Pierre Dupont 22 2002 Grammar Induction 2002 Grammar Induction Merging is fun a 3 A1 a b a 0 1 2 b b a a 1 4 6 8 0 b a a 2 5 7 b a b a a A b a 2 0,1 2 0 1,2 a b 0,1 2 0,2 1 a b a 0 1 b a 0,1,2 b a • Merging ⇔ definition of a partition π on the set of states 0 Example : {{0,1}, {2}} How are we going to find the right partition? Use negative data! • If A2 = A1/π then L(A1) ⊆ L(A2) : merging states ⇔ generalize language Pierre Dupont 21 Pierre Dupont 23 2002 Grammar Induction 2002 Grammar Induction Summary • Grammar induction definition A(L ) 0 Data Grammar Generation Induction PTA π • Learning paradigms PTA π ? • DFA learning from positive and negative examples We observe some positive and negative data • RPNI algorithm The positive sample S+ comes from a regular language L0 • Probabilistic DFA learning The positive sample is assumed to be structurally complete with respect to the • Application to a natural language task canonical automaton A(L0) of the target language L0 (Not an additional hypothesis but a way to restrict the search to reasonable generalizations!) • Links with Markov models • Smoothing issues We build the Prefix Tree Acceptor of S+. By construction L(PTA) = S+ • Related problems and future work Merging states ⇔ generalize S+ The negative sample S− helps to control over-generalization Note: finding the minimal DFA consistent with S+,S− is NP-complete! Pierre Dupont 24 Pierre Dupont 26 2002 Grammar Induction 2002 Grammar Induction An automaton induction algorithm RPNI algorithm RPNI is a particular instance of the “generalization as search” paradigm RPNI follows the prefix order in PTA Algorithm Automaton Induction input a 3 S // positive sample b b a + a 1 4 6 8 S // negative sample − 0 b a a A ← PTA(S+) // PTA 2 5 7 while (i, j) ←choose states () do // Choose a state pair Polynomial time complexity with respect to sample size (S+,S−) if compatible (i, j, S−) then // Check for compatibility of merging i and j A ← A/πij RPNI identifies in the limit the class of regular languages end if A characteristic sample, i.e. a sample such that RPNI is guaranteed to produce end while the correct solution, has a quadratic size with respect to |A(L )| return A 0 Additional heuristics exist to improve performance when such a sample is not provided Pierre Dupont 25 Pierre Dupont 27 2002 Grammar Induction 2002 Grammar Induction RPNI algorithm: pseudo-code Search space characterization input S+,S− Conditions on the learning sample to guarantee the existence of a solution output A DFA consistent with S+,S− begin DFA and NFA in the lattice A ← PTA(S+) // N denotes the number of states of PTA(S+) π ← {{0}, {1},..., {N
Load more

Recommended publications
  • Unsupervised Recurrent Neural Network Grammars
    Unsupervised Recurrent Neural Network Grammars Yoon Kim Alexander Rush Lei Yu Adhiguna Kuncoro Chris Dyer G´abor Melis Code: https://github.com/harvardnlp/urnng 1/36 Language Modeling & Grammar Induction Goal of Language Modeling: assign high likelihood to held-out data Goal of Grammar Induction: learn linguistically meaningful tree structures without supervision Incompatible? For good language modeling performance, need little independence assumptions and make use of flexible models (e.g. deep networks) For grammar induction, need strong independence assumptions for tractable training and to imbue inductive bias (e.g. context-freeness grammars) 2/36 Language Modeling & Grammar Induction Goal of Language Modeling: assign high likelihood to held-out data Goal of Grammar Induction: learn linguistically meaningful tree structures without supervision Incompatible? For good language modeling performance, need little independence assumptions and make use of flexible models (e.g. deep networks) For grammar induction, need strong independence assumptions for tractable training and to imbue inductive bias (e.g. context-freeness grammars) 2/36 This Work: Unsupervised Recurrent Neural Network Grammars Use a flexible generative model without any explicit independence assumptions (RNNG) =) good LM performance Variational inference with a structured inference network (CRF parser) to regularize the posterior =) learn linguistically meaningful trees 3/36 Background: Recurrent Neural Network Grammars [Dyer et al. 2016] Structured joint generative model of sentence x and tree z pθ(x; z) Generate next word conditioned on partially-completed syntax tree Hierarchical generative process (cf. flat generative process of RNN) 4/36 Background: Recurrent Neural Network Language Models Standard RNNLMs: flat left-to-right generation xt ∼ pθ(x j x1; : : : ; xt−1) = softmax(Wht−1 + b) 5/36 Background: RNNG [Dyer et al.
    [Show full text]
  • Deep Learning for Source Code Modeling and Generation: Models, Applications and Challenges
    Deep Learning for Source Code Modeling and Generation: Models, Applications and Challenges TRIET H. M. LE, The University of Adelaide HAO CHEN, The University of Adelaide MUHAMMAD ALI BABAR, The University of Adelaide Deep Learning (DL) techniques for Natural Language Processing have been evolving remarkably fast. Recently, the DL advances in language modeling, machine translation and paragraph understanding are so prominent that the potential of DL in Software Engineering cannot be overlooked, especially in the field of program learning. To facilitate further research and applications of DL in this field, we provide a comprehensive review to categorize and investigate existing DL methods for source code modeling and generation. To address the limitations of the traditional source code models, we formulate common program learning tasks under an encoder-decoder framework. After that, we introduce recent DL mechanisms suitable to solve such problems. Then, we present the state-of-the-art practices and discuss their challenges with some recommendations for practitioners and researchers as well. CCS Concepts: • General and reference → Surveys and overviews; • Computing methodologies → Neural networks; Natural language processing; • Software and its engineering → Software notations and tools; Source code generation; Additional Key Words and Phrases: Deep learning, Big Code, Source code modeling, Source code generation 1 INTRODUCTION Deep Learning (DL) has recently emerged as an important branch of Machine Learning (ML) because of its incredible performance in Computer Vision and Natural Language Processing (NLP) [75]. In the field of NLP, it has been shown that DL models can greatly improve the performance of many classic NLP tasks such as semantic role labeling [96], named entity recognition [209], machine translation [256], and question answering [174].
    [Show full text]
  • GRAINS: Generative Recursive Autoencoders for Indoor Scenes
    GRAINS: Generative Recursive Autoencoders for INdoor Scenes MANYI LI, Shandong University and Simon Fraser University AKSHAY GADI PATIL, Simon Fraser University KAI XU∗, National University of Defense Technology SIDDHARTHA CHAUDHURI, Adobe Research and IIT Bombay OWAIS KHAN, IIT Bombay ARIEL SHAMIR, The Interdisciplinary Center CHANGHE TU, Shandong University BAOQUAN CHEN, Peking University DANIEL COHEN-OR, Tel Aviv University HAO ZHANG, Simon Fraser University Bedrooms Living rooms Kitchen Fig. 1. We present a generative recursive neural network (RvNN) based on a variational autoencoder (VAE) to learn hierarchical scene structures, enabling us to easily generate plausible 3D indoor scenes in large quantities and varieties (see scenes of kitchen, bedroom, office, and living room generated). Using the trained RvNN-VAE, a novel 3D scene can be generated from a random vector drawn from a Gaussian distribution in a fraction of a second. We present a generative neural network which enables us to generate plau- learned distribution. We coin our method GRAINS, for Generative Recursive sible 3D indoor scenes in large quantities and varieties, easily and highly Autoencoders for INdoor Scenes. We demonstrate the capability of GRAINS efficiently. Our key observation is that indoor scene structures are inherently to generate plausible and diverse 3D indoor scenes and compare with ex- hierarchical. Hence, our network is not convolutional; it is a recursive neural isting methods for 3D scene synthesis. We show applications of GRAINS network or RvNN. Using a dataset of annotated scene hierarchies, we train including 3D scene modeling from 2D layouts, scene editing, and semantic a variational recursive autoencoder, or RvNN-VAE, which performs scene scene segmentation via PointNet whose performance is boosted by the large object grouping during its encoding phase and scene generation during quantity and variety of 3D scenes generated by our method.
    [Show full text]
  • Unsupervised Recurrent Neural Network Grammars
    Unsupervised Recurrent Neural Network Grammars Yoon Kimy Alexander M. Rushy Lei Yu3 Adhiguna Kuncoroz;3 Chris Dyer3 Gabor´ Melis3 yHarvard University zUniversity of Oxford 3DeepMind fyoonkim,[email protected] fleiyu,akuncoro,cdyer,[email protected] Abstract The standard setup for unsupervised structure p (x; z) Recurrent neural network grammars (RNNG) learning is to define a generative model θ are generative models of language which over observed data x (e.g. sentence) and unob- jointly model syntax and surface structure by served structure z (e.g. parse tree, part-of-speech incrementally generating a syntax tree and sequence), and maximize the log marginal like- P sentence in a top-down, left-to-right order. lihood log pθ(x) = log z pθ(x; z). Success- Supervised RNNGs achieve strong language ful approaches to unsupervised parsing have made modeling and parsing performance, but re- strong conditional independence assumptions (e.g. quire an annotated corpus of parse trees. In context-freeness) and employed auxiliary objec- this work, we experiment with unsupervised learning of RNNGs. Since directly marginal- tives (Klein and Manning, 2002) or priors (John- izing over the space of latent trees is in- son et al., 2007). These strategies imbue the learn- tractable, we instead apply amortized varia- ing process with inductive biases that guide the tional inference. To maximize the evidence model to discover meaningful structures while al- lower bound, we develop an inference net- lowing tractable algorithms for marginalization; work parameterized as a neural CRF con- however, they come at the expense of language stituency parser. On language modeling, unsu- modeling performance, particularly compared to pervised RNNGs perform as well their super- vised counterparts on benchmarks in English sequential neural models that make no indepen- and Chinese.
    [Show full text]
  • Semantic Analysis of Multi Meaning Words Using Machine Learning and Knowledge Representation by Marjan Alirezaie
    Simpo PDF Merge and Split Unregistered Version - http://www.simpopdf.com Institutionen f¨or datavetenskap Department of Computer and Information Science Final Thesis Semantic Analysis Of Multi Meaning Words Using Machine Learning And Knowledge Representation by Marjan Alirezaie LiU/IDA-EX-A- -11/011- -SE 2011-04-10 Link¨opings universitet Link¨opings universitet SE-581 83 Link¨opings, Sweden 581 83 Link¨opings Simpo PDF Merge and Split Unregistered Version - http://www.simpopdf.com Simpo PDF Merge and Split Unregistered Version - http://www.simpopdf.com Final Thesis Semantic Analysis Of Multi Meaning Words Using Machine Learning And Knowledge Representation by Marjan Alirezaie LiU/IDA-EX-A- -11/011- -SE Supervisor, Examiner : Professor Erik Sandewall Dept. of Computer and Information Science at Link¨opings Universitet Simpo PDF Merge and Split Unregistered Version - http://www.simpopdf.com Simpo PDF Merge and Split Unregistered Version - http://www.simpopdf.com Abstract The present thesis addresses machine learning in a domain of natural- language phrases that are names of universities. It describes two approaches to this problem and a software implementation that has made it possible to evaluate them and to compare them. In general terms, the system's task is to learn to 'understand' the signif- icance of the various components of a university name, such as the city or region where the university is located, the scientific disciplines that are studied there, or the name of a famous person which may be part of the university name. A concrete test for whether the system has acquired this understanding is when it is able to compose a plausible university name given some components that should occur in the name.
    [Show full text]
  • Max-Margin Synchronous Grammar Induction for Machine Translation
    Max-Margin Synchronous Grammar Induction for Machine Translation Xinyan Xiao and Deyi Xiong∗ School of Computer Science and Technology Soochow University Suzhou 215006, China [email protected], [email protected] Abstract is not elegant theoretically. It brings an undesirable gap that separates modeling and learning in an SMT Traditional synchronous grammar induction system. estimates parameters by maximizing likeli- Therefore, researchers have proposed alternative hood, which only has a loose relation to trans- lation quality. Alternatively, we propose a approaches to learning synchronous grammars di- max-margin estimation approach to discrim- rectly from sentence pairs without word alignments, inatively inducing synchronous grammars for via generative models (Marcu and Wong, 2002; machine translation, which directly optimizes Cherry and Lin, 2007; Zhang et al., 2008; DeNero translation quality measured by BLEU. In et al., 2008; Blunsom et al., 2009; Cohn and Blun- the max-margin estimation of parameters, we som, 2009; Neubig et al., 2011; Levenberg et al., only need to calculate Viterbi translations. 2012) or discriminative models (Xiao et al., 2012). This further facilitates the incorporation of Theoretically, these approaches describe how sen- various non-local features that are defined on the target side. We test the effectiveness of our tence pairs are generated by applying sequences of max-margin estimation framework on a com- synchronous rules in an elegant way. However, they petitive hierarchical phrase-based system. Ex- learn synchronous grammars by maximizing likeli- periments show that our max-margin method hood,1 which only has a loose relation to transla- significantly outperforms the traditional two- tion quality (He and Deng, 2012).
    [Show full text]
  • Compound Probabilistic Context-Free Grammars for Grammar Induction
    Compound Probabilistic Context-Free Grammars for Grammar Induction Yoon Kim Chris Dyer Alexander M. Rush Harvard University DeepMind Harvard University Cambridge, MA, USA London, UK Cambridge, MA, USA [email protected] [email protected] [email protected] Abstract non-parametric models (Kurihara and Sato, 2006; Johnson et al., 2007; Liang et al., 2007; Wang and We study a formalization of the grammar in- Blunsom, 2013), and manually-engineered fea- duction problem that models sentences as be- tures (Huang et al., 2012; Golland et al., 2012) to ing generated by a compound probabilistic context free grammar. In contrast to traditional encourage the desired structures to emerge. formulations which learn a single stochastic We revisit these aforementioned issues in light grammar, our context-free rule probabilities of advances in model parameterization and infer- are modulated by a per-sentence continuous ence. First, contrary to common wisdom, we latent variable, which induces marginal de- find that parameterizing a PCFG’s rule probabil- pendencies beyond the traditional context-free ities with neural networks over distributed rep- assumptions. Inference in this grammar is resentations makes it possible to induce linguis- performed by collapsed variational inference, tically meaningful grammars by simply optimiz- in which an amortized variational posterior is placed on the continuous variable, and the la- ing log likelihood. While the optimization prob- tent trees are marginalized with dynamic pro- lem remains non-convex, recent work suggests gramming. Experiments on English and Chi- that there are optimization benefits afforded by nese show the effectiveness of our approach over-parameterized models (Arora et al., 2018; compared to recent state-of-the-art methods Xu et al., 2018; Du et al., 2019), and we in- for grammar induction from words with neu- deed find that this neural PCFG is significantly ral language models.
    [Show full text]
  • Grammar Induction
    Grammar Induction Regina Barzilay MIT October, 2005 Three non-NLP questions 1. Which is the odd number out? 625,361,256,197,144 2. Insert the missing letter: B,E,?,Q,Z 3. Complete the following number sequence: 4, 6, 9, 13 7, 10, 15, ? How do you solve these questions? • Guess a pattern that generates the sequence Insert the missing letter: B,E,?,Q,Z 2,5,?,17, 26 k2 + 1 • Select a solution based on the detected pattern k = 3 ! 10th letter of the alphabet ! J More Patterns to Decipher: Byblos Script Image removed for copyright reasons. More Patterns to Decipher: Lexicon Learning Ourenemiesareinnovativeandresourceful,andsoarewe. Theyneverstopthinkingaboutnewwaystoharmourcountry andourpeople,andneitherdowe. Which is the odd word out? Ourenemies . Enemies . We . More Patterns to Decipher: Natural Language Syntax Which is the odd sentence out? The cat eats tuna. The cat and the dog eats tuna. Today • Vocabulary Induction – Word Boundary Detection • Grammar Induction – Feasibility of language acquisition – Algorithms for grammar induction Vocabulary Induction Task: Unsupervised learning of word boundary segmentation • Simple: Ourenemiesareinnovativeandresourceful,andsoarewe. Theyneverstopthinkingaboutnewwaystoharmourcountry andourpeople,andneitherdowe. • More ambitious: Image of Byblos script removed for copyright reasons. Word Segmentation (Ando&Lee, 2000) Key idea: for each candidate boundary, compare the frequency of the n-grams adjacent to the proposed boundary with the frequency of the n-grams that straddle it. S ? S 1 2 T I N G E V I D t 1 t 2 t3 For N = 4, consider the 6 questions of the form: ”Is #(si ) � #(tj )?”, where #(x) is the number of occurrences of x Example: Is “TING” more frequent in the corpus than ”INGE”? Algorithm for Word Segmentation s n non-straddling n-grams to the left of location k 1 s n non-straddling n-grams to the right of location k 2 tn straddling n-gram with j characters to the right of location k j I� (y; z) indicator function that is 1 when y � z, and 0 otherwise.
    [Show full text]
  • Dependency Grammar Induction with a Neural Variational Transition
    The Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19) Dependency Grammar Induction with a Neural Variational Transition-Based Parser Bowen Li, Jianpeng Cheng, Yang Liu, Frank Keller Institute for Language, Cognition and Computation School of Informatics, University of Edinburgh 10 Crichton Street, Edinburgh EH8 9AB, UK fbowen.li, jianpeng.cheng, [email protected], [email protected] Abstract Though graph-based models achieve impressive results, their inference procedure requires O(n3) time complexity. Dependency grammar induction is the task of learning de- Meanwhile, features in graph-based models must be decom- pendency syntax without annotated training data. Traditional graph-based models with global inference achieve state-of- posable over substructures to enable dynamic programming. the-art results on this task but they require O(n3) run time. In comparison, transition-based models allow faster infer- Transition-based models enable faster inference with O(n) ence with linear time complexity and richer feature sets. time complexity, but their performance still lags behind. Although relying on local inference, transition-based mod- In this work, we propose a neural transition-based parser els have been shown to perform well in supervised parsing for dependency grammar induction, whose inference proce- (Kiperwasser and Goldberg 2016; Dyer et al. 2015). How- dure utilizes rich neural features with O(n) time complex- ever, unsupervised transition parsers are not well-studied. ity. We train the parser with an integration of variational in- One exception is the work of Rasooli and Faili (2012), in ference, posterior regularization and variance reduction tech- which search-based structure prediction (Daume´ III 2009) is niques.
    [Show full text]
  • Are Pre-Trained Language Models Aware of Phrases?Simplebut Strong Baselinesfor Grammar Induction
    Published as a conference paper at ICLR 2020 ARE PRE-TRAINED LANGUAGE MODELS AWARE OF PHRASES?SIMPLE BUT STRONG BASELINES FOR GRAMMAR INDUCTION Taeuk Kim1, Jihun Choi1, Daniel Edmiston2 & Sang-goo Lee1 1Dept. of Computer Science and Engineering, Seoul National University, Seoul, Korea 2Dept. of Linguistics, University of Chicago, Chicago, IL, USA ftaeuk,jhchoi,[email protected], [email protected] ABSTRACT With the recent success and popularity of pre-trained language models (LMs) in natural language processing, there has been a rise in efforts to understand their in- ner workings. In line with such interest, we propose a novel method that assists us in investigating the extent to which pre-trained LMs capture the syntactic notion of constituency. Our method provides an effective way of extracting constituency trees from the pre-trained LMs without training. In addition, we report intriguing findings in the induced trees, including the fact that some pre-trained LMs outper- form other approaches in correctly demarcating adverb phrases in sentences. 1 INTRODUCTION Grammar induction, which is closely related to unsupervised parsing and latent tree learning, allows one to associate syntactic trees, i.e., constituency and dependency trees, with sentences. As gram- mar induction essentially assumes no supervision from gold-standard syntactic trees, the existing approaches for this task mainly rely on unsupervised objectives, such as language modeling (Shen et al., 2018b; 2019; Kim et al., 2019a;b) and cloze-style word prediction (Drozdov et al., 2019) to train their task-oriented models. On the other hand, there is a trend in the natural language pro- cessing (NLP) community of leveraging pre-trained language models (LMs), e.g., ELMo (Peters et al., 2018) and BERT (Devlin et al., 2019), as a means of acquiring contextualized word represen- tations.
    [Show full text]
  • What Do Recurrent Neural Network Grammars Learn About Syntax?
    What Do Recurrent Neural Network Grammars Learn About Syntax? Adhiguna Kuncoro♠ Miguel Ballesteros} Lingpeng Kong♠ Chris Dyer♠♣ Graham Neubig♠ Noah A. Smith~ ♠School of Computer Science, Carnegie Mellon University, Pittsburgh, PA, USA }IBM T.J. Watson Research Center, Yorktown Heights, NY, USA |DeepMind, London, UK ~Computer Science & Engineering, University of Washington, Seattle, WA, USA fakuncoro,lingpenk,[email protected] [email protected], [email protected], [email protected] Abstract data. In some sense, such models can be thought of as mini-scientists. Recurrent neural network grammars Neural networks, including RNNGs, are capa- (RNNG) are a recently proposed prob- ble of representing larger classes of hypotheses abilistic generative modeling family for than traditional probabilistic models, giving them natural language. They show state-of- more freedom to explore. Unfortunately, they tend the-art language modeling and parsing to be bad mini-scientists, because their parameters performance. We investigate what in- are difficult for human scientists to interpret. formation they learn, from a linguistic RNNGs are striking because they obtain state- perspective, through various ablations of-the-art parsing and language modeling perfor- to the model and the data, and by aug- mance. Their relative lack of independence as- menting the model with an attention sumptions, while still incorporating a degree of mechanism (GA-RNNG) to enable closer linguistically-motivated prior knowledge, affords inspection. We find that explicit modeling the model considerable freedom to derive its own of composition is crucial for achieving the insights about syntax. If they are mini-scientists, best performance. Through the attention the discoveries they make should be of particular mechanism, we find that headedness interest as propositions about syntax (at least for plays a central role in phrasal represen- the particular genre and dialect of the data).
    [Show full text]
  • A Survey of Unsupervised Dependency Parsing
    A Survey of Unsupervised Dependency Parsing Wenjuan Han1,∗ Yong Jiang2,∗ Hwee Tou Ng1, Kewei Tu3† 1Department of Computer Science, National University of Singapore 2Alibaba DAMO Academy, Alibaba Group 3School of Information Science and Technology, ShanghaiTech University [email protected] [email protected] [email protected] [email protected] Abstract Syntactic dependency parsing is an important task in natural language processing. Unsupervised dependency parsing aims to learn a dependency parser from sentences that have no annotation of their correct parse trees. Despite its difficulty, unsupervised parsing is an interesting research direction because of its capability of utilizing almost unlimited unannotated text data. It also serves as the basis for other research in low-resource parsing. In this paper, we survey existing approaches to unsupervised dependency parsing, identify two major classes of approaches, and discuss recent trends. We hope that our survey can provide insights for researchers and facilitate future research on this topic. 1 Introduction Dependency parsing is an important task in natural language processing that aims to capture syntactic information in sentences in the form of dependency relations between words. It finds applications in semantic parsing, machine translation, relation extraction, and many other tasks. Supervised learning is the main technique used to automatically learn a dependency parser from data. It requires the training sentences to be manually annotated with their correct parse trees. Such a training dataset is called a treebank. A major challenge faced by supervised learning is that treebanks are not al- ways available for new languages or new domains and building a high-quality treebank is very expensive and time-consuming.
    [Show full text]