NTL – Converging Constraints • Basic concepts and words derive their meaning from embodied experience. • Abstract and theoretical concepts derive their meaning from metaphorical maps to more basic embodied concepts. • Structured Connectionist Models can capture both of these processes nicely. • Grammar extends this by Constructions: pairings of form with embodied meaning. Simulation-based language understanding “Harry walked to the cafe.” Utterance Constructions Analysis Process General Knowledge Belief State Schema walk Trajector Harry Cafe Goal cafe Simulation Specification Simulation The ICSI/Berkeley Neural Theory of Language Project Background: Primate Motor Control • Relevant requirements (Stromberg, Latash, Kandel, Arbib, Jeannerod, Rizzolatti) – Should model coordinated, distributed, parameterized control programs required for motor action and perception. – Should be an active structure. – Should be able to model concurrent actions and interrupts. • Model – The NTL project has developed a computational model based on that satisfies these requirements (x- schemas). – Details, papers, etc. can be obtained on the web at http://www.icsi.berkeley.edu/NTL Active representations • Representation based on stochastic Petri nets captures dynamic, parameterized nature of actions • Many inferences about actions derive from what we know about executing them • Generative model: action, planning, recognition, language. walker at goal energy walker=Harry goal=home Walking: bound to a specific walker with a direction or goal consumes resources (e.g., energy) may have termination condition (e.g., walker at goal) ongoing, iterative action Active Motion Model Evolving Responses of Competing Models over Time. Nigel Goddard 1989 Language Development in Children • • • • • • • • 0-3 mo: prefers sounds in native language 3-6 mo: imitation of vowel sounds only 6-8 mo: babbling in consonant-vowel segments 8-10 mo: word comprehension, starts to lose sensitivity to consonants outside native language 12-13 mo: word production (naming) 16-20 mo: word combinations, relational words (verbs, adj.) 24-36 mo: grammaticization, inflectional morphology 3 years – adulthood: vocab. growth, sentence-level grammar for discourse purposes Learning Spatial Relation Words Terry Regier A model of children learning spatial relations. Assumes child hears one word label of scene. Program learns well enough to label novel scenes correctly. Extended to simple motion scenarios, like INTO. System works across languages. Mechanisms are neurally plausible. Learning System dynamic relations (e.g. into) structured connectionist network (based on visual system) We’ll look at the details next lecture Limitations • • • • • • • Scale Uniqueness/Plausibility Grammar Abstract Concepts Inference Representation Biological Realism Constrained Best Fit in Nature inanimate physics chemistry biology vision language animate lowest energy state molecular minima fitness, MEU neuroeconomics threats, friends errors, NTL Learning Verb Meanings David Bailey A model of children learning their first verbs. Assumes parent labels child’s actions. Child knows parameters of action, associates with word Program learns well enough to: 1) Label novel actions correctly 2) Obey commands using new words (simulation) System works across languages Mechanisms are neurally plausible. Motor Control (X-schema) for SLIDE Parameters for the SLIDE X-schema Feature Structures for PUSH System Overview Learning Two Senses of PUSH Model merging based on Bayesian MDL Training Results David Bailey English • 165 Training Examples, 18 verbs • Learns optimal number of word senses (21) • 32 Test examples : 78% recognition, 81% action • All mistakes were close lift ~ yank, etc. • Learned some particle CXN,e.g., pull up Farsi • With identical settings, learned senses not in English Constrained Best Fit in Nature inanimate physics chemistry biology vision language animate lowest energy state molecular minima fitness, MEU neuroeconomics threats, friends errors, NTL Compositionality • Traditional Context-free composition of logical forms • Contemporary Constructional composition of conceptual frames Formal Cognitive Linguistics Embodied Construction Grammar (Bergen, Chang & Paskin 2000) • Assumptions from Construction Grammar – Constructions are form-meaning pairs (Lakoff 1987, Goldberg 1995) – Constructions vary in degree of specificity and level of description (morphological, lexical, phrasal, clausal) • Constructions evoke and bind semantic schemas • Additional influences – Cognitive Grammar (Langacker 1987) – Frame Semantics (Fillmore 1982) – Structured Connectionism (Feldman 1987) Traditional Levels of Analysis Pragmatics Semantics Syntax Morphology Phonetics “Harry walked into the cafe.” Pragmatics Semantics Utterance Syntax Morphology Phonetics Language understanding: analysis & simulation construction WALKED in context c constituents: form f of type [wakt] meaning walking construed as Walk-Action semantic constraints: walking.time before c.speech-time walking.aspect = encapsulated designates walking “Harry walked into the cafe.” Utterance Analysis Process Constructions General Knowledge Simulation Specification Belief State CAFE Simulation Simulation specification A simulation specification consists of: •semantic schemas evoked by constructions •bindings between schemas (labeled by the constructions that enforce them) Conclusion • Language acquisition and use is a hallmark of being human – Language seems to rely on fine-grained aspects of embodied (sensory-motor and social cognition) primitives and brain-like computation (massively parallel, distributed, spreading activation, temporal binding). – Understanding requires imaginative simulation! – We have built a pilot system that demonstrates the use of motor control representations in grounding the language of abstract actions and policies. • Sensory-Motor imagination and simulation is crucial in interpretation! • Ongoing Work. – Formalize and use a compositional set of embodied conceptual primitives and grammatical constructions. – Perform both behavioral and fMRI imaging experiments to test the predictions of the simulation hypothesis. – Further refine and ground the model in details of neural anatomy and functional architecture (basal-thalamic-cortical loops).