We present a generative distributional model for the unsupervised induction of natural language syntax which explicitly models constituent yields and contexts. Parameter search with EM produces higher quality analyses than previously exhibited by unsupervised systems, giving the best published unsupervised parsing results on the ATIS corpus. Experiments on Penn treebank sentences of comparable length show an even higher F1 of 71% on nontrivial brackets. We compare distributionally induced and actual part-of-speech tags as input data, and examine (...) extensions to the basic model. We discuss errors made by the system, compare the system to previous models, and discuss upper bounds, lower bounds, and stability for this task. (shrink)

How should a machine intelligence perform unsupervised structure discovery over streams of sensory input? One approach to this problem is to cast it as an apperception task [1]. Here, the task is to construct an explicit interpretable theory that both explains the sensory sequence and also satisfies a set of unity conditions, designed to ensure that the constituents of the theory are connected in a relational structure. However, the original formulation of the apperception task had one fundamental limitation: it (...) assumed the raw sensory input had already been parsed using a set of discrete categories, so that all the system had to do was receive this already-digested symbolic input, and make sense of it. But what if we don't have access to pre-parsed input? What if our sensory sequence is raw unprocessed information? The central contribution of this paper is a neuro-symbolic framework for distilling interpretable theories out of streams of raw, unprocessed sensory experience. First, we extend the definition of the apperception task to include ambiguous (but still symbolic) input: sequences of sets of disjunctions. Next, we use a neural network to map raw sensory input to disjunctive input. Our binary neural network is encoded as a logic program, so the weights of the network and the rules of the theory can be solved jointly as a single SAT problem. This way, we are able to jointly learn how to perceive (mapping raw sensory information to concepts) and apperceive (combining concepts into declarative rules). (shrink)

While rules and exemplars are usually viewed as opposites, this paper argues that they form end points of the same distribution. By representing both rules and exemplars as (partial) trees, we can take into account the fluid middle ground between the two extremes. This insight is the starting point for a new theory of language learning that is based on the following idea: If a language learner does not know which phrase‐structure trees should be assigned to initial sentences, s/he allows (...) (implicitly) for all possible trees and lets linguistic experience decide which is the “best” tree for each sentence. The best tree is obtained by maximizing “structural analogy” between a sentence and previous sentences, which is formalized by the most probable shortest combination of subtrees from all trees of previous sentences. Corpus‐based experiments with this model on the Penn Treebank and the Childes database indicate that it can learn both exemplar‐based and rule‐based aspects of language, ranging from phrasal verbs to auxiliary fronting. By having learned the syntactic structures of sentences, we have also learned the grammar implicit in these structures, which can in turn be used to produce new sentences. We show that our model mimicks children’s language development from item‐based constructions to abstract constructions, and that the model can simulate some of the errors made by children in producing complex questions. (shrink)

We introduce a set of biologically and computationally motivated design choices for modeling the learning of language, or of other types of sequential, hierarchically structured experience and behavior, and describe an implemented system that conforms to these choices and is capable of unsupervised learning from raw natural-language corpora. Given a stream of linguistic input, our model incrementally learns a grammar that captures its statistical patterns, which can then be used to parse or generate new data. The grammar constructed in (...) this manner takes the form of a directed weighted graph, whose nodes are recursively (hierarchically) defined patterns over the elements of the input stream. We evaluated the model in seventeen experiments, grouped into five studies, which examined, respectively, (a) the generative ability of grammar learned from a corpus of natural language, (b) the characteristics of the learned representation, (c) sequence segmentation and chunking, (d) artificial grammar learning, and (e) certain types of structure dependence. The model's performance largely vindicates our design choices, suggesting that progress in modeling language acquisition can be made on a broad front—ranging from issues of generativity to the replication of human experimental findings—by bringing biological and computational considerations, as well as lessons from prior efforts, to bear on the modeling approach. (shrink)

An important element of Natural Language Processing is parts of speech tagging. With fine-grained word-class annotations, the word forms in a text can be enhanced and can also be used in downstream processes, such as dependency parsing. The improved search options that tagged data offers also greatly benefit linguists and lexicographers. Natural language processing research is becoming increasingly popular and important as unsupervised learning methods are developed. There are some aspects of the Albanian language that make the creation (...) of a part-of-speech tag set challenging. This research provides a discussion of those issues linguistic phenomena and presents a proposal for a part-of-speech tag set that can adequately represent them. The corpus contains more than 250,000 tokens, each annotated with a medium-sized tag set. The Albanian language’s syntagmatic aspects are adequately represented. Additionally, in this paper are morphologically and part-of-speech tagged corpora for the Albanian language, as well as lemmatize and neural morphological tagger trained on these corpora. Based on the held-out evaluation set, the model achieves 93.65% accuracy on part-of-speech tagging, The morphological tagging rate was 85.31 % and the lemmatization rate was 88.95%. Furthermore, the TF-IDF technique weighs terms and with the scores are highlighted words that have additional information for the Albanian corpus. (shrink)

La literatura sobre las relaciones entre las personas y la técnica se refiere a análisis heredados de la economía, la sociología, la política, etc. Estas miradas abarcan a la técnica de forma global dificultando el estudio de algunos procesos que ocurren entre cada persona y la técnica. En este sentido, buscaremos explicitar estas relaciones a través de los productos de la técnica como mediadores sociales. Con esta base tienen sentido los criterios de las tecnologías entrañables para que esta mediación se (...) transforme en una relación deseable en términos sociales, proponiendo a la honestidad como concepto de consenso primario. (shrink)

Modeling law search and retrieval as prediction problems has recently emerged as a predominant approach in law intelligence. Focusing on the law article retrieval task, we present a deep learning framework named LamBERTa, which is designed for civil-law codes, and specifically trained on the Italian civil code. To our knowledge, this is the first study proposing an advanced approach to law article prediction for the Italian legal system based on a BERT (Bidirectional Encoder Representations from Transformers) learning framework, which has (...) recently attracted increased attention among deep learning approaches, showing outstanding effectiveness in several natural language processing and learning tasks. We define LamBERTa models by fine-tuning an Italian pre-trained BERT on the Italian civil code or its portions, for law article retrieval as a classification task. One key aspect of our LamBERTa framework is that we conceived it to address an extreme classification scenario, which is characterized by a high number of classes, the few-shot learning problem, and the lack of test query benchmarks for Italian legal prediction tasks. To solve such issues, we define different methods for the unsupervised labeling of the law articles, which can in principle be applied to any law article code system. We provide insights into the explainability and interpretability of our LamBERTa models, and we present an extensive experimental analysis over query sets of different type, for single-label as well as multi-label evaluation tasks. Empirical evidence has shown the effectiveness of LamBERTa, and also its superiority against widely used deep-learning text classifiers and a few-shot learner conceived for an attribute-aware prediction task. (shrink)

We describe a way of modeling high‐dimensional data vectors by using an unsupervised, nonlinear, multilayer neural network in which the activity of each neuron‐like unit makes an additive contribution to a global energy score that indicates how surprised the network is by the data vector. The connection weights that determine how the activity of each unit depends on the activities in earlier layers are learned by minimizing the energy assigned to data vectors that are actually observed and maximizing the (...) energy assigned to “confabulations” that are generated by perturbing an observed data vector in a direction that decreases its energy under the current model. (shrink)

Artificial Intelligence in the form of different machine learning models is applied to Big Data as a way to turn data into valuable knowledge. The rhetoric is that ensuing predictions work well—with a high degree of autonomy and automation. We argue that we need to analyze the process of applying machine learning in depth and highlight at what point human knowledge production takes place in seemingly autonomous work. This article reintroduces classification theory as an important framework for understanding such seemingly (...) invisible knowledge production in the machine learning development and design processes. We suggest a framework for studying such classification closely tied to different steps in the work process and exemplify the framework on two experiments with machine learning applied to Facebook data from one of our labs. By doing so we demonstrate ways in which classification and potential discrimination take place in even seemingly unsupervised and autonomous models. Moving away from concepts of non-supervision and autonomy enable us to understand the underlying classificatory dispositifs in the work process and that this form of analysis constitutes a first step towards governance of artificial intelligence. (shrink)

We describe a pattern acquisition algorithm that learns, in an unsupervised fashion, a streamlined representation of linguistic structures from a plain natural-language corpus. This paper addresses the issues of learning structured knowledge from a large-scale natural language data set, and of generalization to unseen text. The implemented algorithm represents sentences as paths on a graph whose vertices are words. Significant patterns, determined by recursive context-sensitive statistical inference, form new vertices. Linguistic constructions are represented by trees composed of significant patterns (...) and their associated equivalence classes. An input module allows the algorithm to be subjected to a standard test of English as a Second Language profi- ciency. The results are encouraging: the model attains a level of performance considered to be “intermediate” for 9th-grade students, despite having been trained on a corpus containing transcribed speech of parents directed to small children. (shrink)

Unsupervised learning algorithms are widely used for many important statistical tasks with numerous applications in science and industry. Yet despite their prevalence, they have attracted remarkably little philosophical scrutiny to date. This stands in stark contrast to supervised and reinforcement learning algorithms, which have been widely studied and critically evaluated, often with an emphasis on ethical concerns. In this article, I analyze three canonical unsupervised learning problems: clustering, abstraction, and generative modeling. I argue that these methods raise unique (...) epistemological and ontological questions, providing data-driven tools for discovering natural kinds and distinguishing essence from contingency. This analysis goes some way toward filling the lacuna in contemporary philosophical discourse on unsupervised learning, as well as bringing conceptual unity to a heterogeneous field more often described by what it isnot(i.e., supervised or reinforcement learning) than by what itis. I submit that unsupervised learning is not just a legitimate subject of philosophical inquiry but perhaps the most fundamental branch of all AI. However, an uncritical overreliance on unsupervised methods poses major epistemic and ethical risks. I conclude by advocating for a pragmatic, error-statistical approach that embraces the opportunities and mitigates the challenges posed by this powerful class of algorithms. (shrink)

In this paper I tackle the question of what basic form an analytical method for articulating and ultimately assessing visual representations should take. I start from the assumption that scientific images, being less prone to interpretive complication than artworks, are ideal objects from which to engage this question. I then assess a recent application of Nelson Goodman's aesthetics to the project of parsing scientific images, Laura Perini's ‘The truth in pictures’. I argue that, although her project is an important (...) one, her Goodmanian conventionalism produces a method of analysis that is incapable of adequately parsing a certain class of pictures and her focus on truth is unnecessary. This speaks against the promise of Goodman's analytical strategy for elucidating visual content and reasoning in the sciences and elsewhere. As an alternative, I develop John Willats’ analytical method and compare it to Perini's through engaging three of her examples—a chemical diagram, a graph and an electron micrograph. Ultimately, a space remains open for a mixed system where Willats’ account provides pictorial analysis and the Goodman–Perini approach parses visual languages. (shrink)

Unsupervised statistical learning is the standard setting for the development of the only advanced visual system that is both highly sophisticated and versatile, and extensively studied: that of monkeys and humans. In this extended abstract, we invoke philosophical observations, computational arguments, behavioral data and neurobiological findings to explain why computer vision researchers should care about (1) unsupervised learning, (2) statistical inference, and (3) the visual brain. We then outline a neuromorphic approach to structural primitive learning motivated by these (...) considerations, survey a range of neurobiological findings and behavioral data consistent with it, and conclude by mentioning some of the more challenging directions for future research. (shrink)

Parsing of argumentative structures has become a very active line of research in recent years. Like discourse parsing or any other natural language task that requires prediction of linguistic struc...

We describe a linguistic pattern acquisition algorithm that learns, in an unsupervised fashion, a streamlined representation of corpus data. This is achieved by compactly coding recursively structured constituent patterns, and by placing strings that have an identical backbone and similar context structure into the same equivalence class. The resulting representations constitute an efficient encoding of linguistic knowledge and support systematic generalization to unseen sentences.

To learn a visual code in an unsupervised manner, one may attempt to capture those features of the stimulus set that would contribute significantly to a statistically efficient representation. Paradoxically, all the candidate features in this approach need to be known before statistics over them can be computed. This paradox may be circumvented by confining the repertoire of candidate features to actual scene fragments, which resemble the “what+where” receptive fields found in the ventral visual stream in primates. We describe (...) a single-layer network that learns such fragments from unsegmented raw images of structured objects. The learning method combines fast imprinting in the feedforward stream with lateral interactions to achieve single-epoch unsupervised acquisition of spatially localized features that can support systematic treatment of structured objects [1]. (shrink)

Contemporary Artificial Intelligence methods, especially their subset-deep learning, are finding their way to successful implementations in the detection and classification of intrusions at the network level. This paper presents an intrusion detection mechanism that leverages Deep AutoEncoder and several Deep Decoders for unsupervised classification. This work incorporates multiple network topology setups for comparative studies. The efficiency of the proposed topologies is validated on two established benchmark datasets: UNSW-NB15 and NetML-2020. The results of their analysis are discussed in terms of (...) classification accuracy, detection rate, false-positive rate, negative predictive value, Matthews correlation coefficient and F1-score. Furthermore, comparing against the state-of-the-art methods used for network intrusion detection is also disclosed. (shrink)

This paper contains the full code of an implementation in Haskell [2], in ‘literate programming’ style [3], of an approach to deductive parsing based on [4]. We focus on the case of the Earley [1] parsing algorithm for CF languages.

This paper extends the Earley parsing algorithm for context free languages [3] to the case of sequentially indexed languages. Sequentially indexed languages are related to indexed languages [1, 2]. The difference is that parallel processing of index stacks is replaced by sequential processing [4].

In this paper, we use antecedent-final conditionals to formulate two problems for parsing-based theories of presupposition projection and triviality of the kind given in Schlenker 2009. We show that, when it comes to antecedent-final conditionals, parsing-based theories predict filtering of presuppositions where there is in fact projection, and triviality judgments for sentences which are in fact felicitous. More concretely, these theories predict that presuppositions triggered in the antecedent of antecedent-final conditionals will be filtered (i.e. will not project) if (...) the negation of the consequent entails the presupposition. But this is wrong: John isn’t in Paris, if he regrets being in France intuitively presupposes that John is in France, contrary to this prediction. Likewise, parsing-based approaches to triviality predict that material entailed by the negation of the consequent will be redundant in the antecedent of the conditional; but John isn’t in Paris, if he’s in France and Mary is with him is intuitively felicitous, contrary to these predictions. Importantly, given that the trigger appears in sentence-final position, both incremental (left-to-right) and symmetric versions of such theories make the same predictions. These data constitute a challenge to the idea that presupposition projection and triviality should be computed on the basis of parsing. This issue is important because it relates to the more general question as to whether presupposition and triviality calculation should be thought of as a pragmatic post-compositional phenomenon or as part of compositional semantics (as in the more traditional dynamic approaches). We discuss a solution which allows us to maintain the parsing-based pragmatic approach; it is based on an analysis of conditionals which incorporates a presupposition that their antecedent is compatible with the context, together with a modification to Schlenker’s (2009) algorithm for calculating local contexts so that it takes into account presupposed material. As we will discuss, this solution works within a framework broadly similar to that of Schlenker’s (2009), but it doesn’t extend in an obvious way to other parsing-based accounts (e.g. parsing-based trivalent approaches). We conclude that a parsing-based theory can be maintained, but only if we adopt a substantial change of perspective on the framework. (shrink)

The automated identification of national implementations of European directives by text similarity techniques has shown promising preliminary results. Previous works have proposed and utilized unsupervised lexical and semantic similarity techniques based on vector space models, latent semantic analysis and topic models. However, these techniques were evaluated on a small multilingual corpus of directives and NIMs. In this paper, we utilize word and paragraph embedding models learned by shallow neural networks from a multilingual legal corpus of European directives and national (...) legislation to develop unsupervised semantic similarity systems to identify transpositions. We evaluate these models and compare their results with the previous unsupervised methods on a multilingual test corpus of 43 Directives and their corresponding NIMs. We also develop supervised machine learning models to identify transpositions and compare their performance with different feature sets. (shrink)

This paper develops a novel psycholinguistic parser and tests it against experimental and corpus reading data. The parser builds on the recent research into memory structures, which argues that memory retrieval is content‐addressable and cue‐based. It is shown that the theory of cue‐based memory systems can be combined with transition‐based parsing to produce a parser that, when combined with the cognitive architecture ACT‐R, can model reading and predict online behavioral measures (reading times and regressions). The parser's modeling capacities are (...) tested against self‐paced reading experimental data (Grodner & Gibson, 2005), eye‐tracking experimental data (Staub, 2011), and a self‐paced reading corpus (Futrell et al., 2018). (shrink)

The main result in this paper is a method for obtaining derivation trees from sentences of certain formal grammars. No parsing algorithm was previously known to exist for these grammars.Applied to Montague's PTQ the method produces all parses that could correspond to different meanings. The technique directly addresses scope and reference and provides a framework for examining these phenomena. The solution for PTQ is implemented in an efficient and useful computer program.

This report details experimental results of using stochastic disambiguation models for parsing sentences from the Redwoods treebank (Oepen et al., 2002). The goals of this paper are two-fold: (i) to report accuracy results on the more highly ambiguous latest version of the treebank, as compared to already published results achieved by the same stochastic models on a previous version of the corpus, and (ii) to present some newly developed models using features from the HPSG signs, as well as the (...) MRS dependency graphs. (shrink)

This paper presents empirical studies and closely corresponding theoretical models of the performance of a chart parser exhaustively parsing the Penn Treebank with the Treebank’s own CFG grammar. We show how performance is dramatically affected by rule representation and tree transformations, but little by top-down vs. bottom-up strategies. We discuss grammatical saturation, including analysis of the strongly connected components of the phrasal nonterminals in the Treebank, and model how, as sentence length increases, the effective grammar rule size increases as (...) regions of the grammar are unlocked, yielding super-cubic observed time behavior in some configurations. (shrink)

We address the problem of predicting how people will spontaneously divide into groups a set of novel items. This is a process akin to perceptual organization. We therefore employ the simplicity principle from perceptual organization to propose a simplicity model of unconstrained spontaneous grouping. The simplicity model predicts that people would prefer the categories for a set of novel items that provide the simplest encoding of these items. Classification predictions are derived from the model without information either about the number (...) of categories sought or information about the distributional properties of the objects to be classified. These features of the simplicity model distinguish it from other models in unsupervised categorization (where, for example, the number of categories sought is determined via a free parameter), and we discuss how these computational differences are related to differences in modeling objectives. The predictions of the simplicity model are validated in four experiments. We also discuss the significance of simplicity in cognitive modeling more generally. (shrink)

A* PCFG parsing can dramatically reduce the time required to find the exact Viterbi parse by conservatively estimating outside Viterbi probabilities. We discuss various estimates and give efficient algorithms for computing them. On Penn treebank sentences, our most detailed estimate reduces the total number of edges processed to less than 3% of that required by exhaustive parsing, and even a simpler estimate which can be pre-computed in under a minute still reduces the work by a factor of 5. (...) The algorithm extends the classic A* graph search procedure to a certain hypergraph associated with parsing. Unlike bestfirst and finite-beam methods for achieving this kind of speed-up, the A* parser is guaranteed to return the most likely parse, not just an approximation. The algorithm is also correct for a wide range of parser control strategies and maintains a worst-case cubic time bound. (shrink)

tic structure. Determining the semantic roles of a verb’s dependents is an important step in natural..

While symbolic parsers can be viewed as deduction systems, this view is less natural for probabilistic parsers. We present a view of parsing as directed hypergraph analysis which naturally covers both symbolic and probabilistic parsing. We illustrate the approach by showing how a dynamic extension of Dijkstra’s algorithm can be used to construct a probabilistic chart parser with an Ç´Ò¿µ time bound for arbitrary PCFGs, while preserving as much of the flexibility of symbolic chart parsers as allowed by (...) the inherent ordering of probabilistic dependencies. (shrink)

Music can be interpreted by attributing syntactic relationships to sequential musical events, and, computationally, such musical interpretation represents an analogous combinatorial task to syntactic processing in language. While this perspective has been primarily addressed in the domain of harmony, we focus here on rhythm in the Western tonal idiom, and we propose for the first time a framework for modeling the moment‐by‐moment execution of processing operations involved in the interpretation of music. Our approach is based on (1) a music‐theoretically motivated (...) grammar formalizing the competence of rhythmic interpretation in terms of three basic types of dependency (preparation, syncopation, and split; Rohrmeier, 2020), and (2) psychologically plausible predictions about the complexity of structural integration and memory storage operations, necessary for parsing hierarchical dependencies, derived from the dependency locality theory (Gibson, 2000). With a behavioral experiment, we exemplify an empirical implementation of the proposed theoretical framework. One hundred listeners were asked to reproduce the location of a visual flash presented while listening to three rhythmic excerpts, each exemplifying a different interpretation under the formal grammar. The hypothesized execution of syntactic‐processing operations was found to be a significant predictor of the observed displacement between the reported and the objective location of the flashes. Overall, this study presents a theoretical approach and a first empirical proof‐of‐concept for modeling the cognitive process resulting in such interpretation as a form of syntactic parsing with algorithmic similarities to its linguistic counterpart. Results from the present small‐scale experiment should not be read as a final test of the theory, but they are consistent with the theoretical predictions after controlling for several possible confounding factors and may form the basis for further large‐scale and ecological testing. (shrink)

We demonstrate that an unlexicalized PCFG can parse much more accurately than previously shown, by making use of simple, linguistically motivated state splits, which break down false independence assumptions latent in a vanilla treebank grammar. Indeed, its performance of 86.36% (LP/LR F1) is better than that of early lexicalized PCFG models, and surprisingly close to the current state-of-theart. This result has potential uses beyond establishing a strong lower bound on the maximum possible accuracy of unlexicalized models: an unlexicalized PCFG is (...) much more compact, easier to replicate, and easier to interpret than more complex lexical models, and the parsing algorithms are simpler, more widely understood, of lower asymptotic complexity, and easier to optimize. (shrink)

In the domain of legal information retrieval, an important challenge is to compute similarity between two legal documents. Precedents play an important role in The Common Law system, where lawyers need to frequently refer to relevant prior cases. Measuring document similarity is one of the most crucial aspects of any document retrieval system which decides the speed, scalability and accuracy of the system. Text-based and network-based methods for computing similarity among case reports have already been proposed in prior works but (...) not without a few pitfalls. Since legal citation networks are generally highly disconnected, network based metrics are not suited for them. Till date, only a few text-based and predominant embedding based methods have been employed, for instance, TF-IDF based approaches, Word2Vec and Doc2Vec based approaches. We investigate the performance of 56 different methodologies for computing textual similarity across court case statements when applied on a dataset of Indian Supreme Court Cases. Among the 56 different methods, thirty are adaptations of existing methods and twenty-six are our proposed methods. The methods studied include models such as BERT and Law2Vec. It is observed that the more traditional methods that rely on a bag-of-words representation performs better than the more advanced context-aware methods for computing document-level similarity. Finally we nominate, via empirical validation, five of our best performing methods as appropriate for measuring similarity between case reports. Among these five, two are adaptations of existing methods and the other three are our proposed methods. (shrink)

In this chapter we consider unsupervised learning from two perspectives. First, we briefly look at its advantages and disadvantages as an engineering technique applied to large corpora in natural language processing. While supervised learning generally achieves greater accuracy with less data, unsupervised learning offers significant savings in the intensive labour required for annotating text. Second, we discuss the possible relevance of unsupervised learning to debates on the cognitive basis of human language acquisition. In this context we explore (...) the implications of recent work on grammar induction for poverty of stimulus arguments that purport to motivate a strong bias model of language learning, commonly formulated as a theory of Universal Grammar (UG). We examine the second issue both as a problem in computational learning theory, and with reference to empirical work on unsupervised Machine Learning (ML) of syntactic structure. We compare two models of learning theory and the place of unsupervised learning within each of them. Looking at recent work on part of speech tagging and the recognition of syntactic structure, we see how far unsupervised ML methods have come in acquiring different kinds of grammatical knowledge from raw text. (shrink)

The University of Pennsylvania Linguistics Department is home of a long-running project to create syntactically annotated (parsed) corpora of historical English. The project is directed by Anthony Kroch, Professor of Linguistics, and the research associate in charge of corpus annotation is Dr. Beatrice Santorini. The Middle English corpus was constructed by Dr. Ann Taylor, now research associate in charge of corpus annotation at the University of York, England.

Scientists increasingly approach the world through machine learning techniques, but philosophers of science often question their epistemic status. Some philosophers have argued that the use of unsupervised clustering algorithms is more justified than the use of supervised classification, because supervised classification is more biased, and because (parametric) simplicity plays a different and more interesting role in unsupervised clustering. I call these arguments the No-Bias Argument and the Simplicity-Truth Argument. I show how both arguments are fallacious and how, on (...) the contrary, the use of supervised classification is at least as justified as the use of unsupervised clustering. (shrink)

We outline an unsupervised language acquisition algorithm and offer some psycholinguistic support for a model based on it. Our approach resembles the Construction Grammar in its general philosophy, and the Tree Adjoining Grammar in its computational characteristics. The model is trained on a corpus of transcribed child-directed speech (CHILDES). The model’s ability to process novel inputs makes it capable of taking various standard tests of English that rely on forced-choice judgment and on magnitude estimation of linguistic acceptability. We report (...) encouraging results from several such tests, and discuss the limitations revealed by other tests in our present method of dealing with novel stimuli. (shrink)