The New Sesquipedalian

The SymSys Forum presents M.S. Project Presentations by Jonathan Tyler Boyd-Meredith and Miriam Connor. Join them on Monday, 5/13 in the Greenberg Room from 12:15-1:05 to hear the following talks:

Detecting Long-Term, Autobiographical Memories Using fMRI (Jonathan Tyler Boyd-Meredith, advised by Anthony Wagner, Psychology)
Machine learning techniques are being applied increasingly to the field of neuroscience to interpret and make use of the data sets generated by fMRI experiments. This has led to striking results for both basic and applied research in many subfields of neuroscience, including learning and memory. In particular, there has been preliminary success in classifying previously encountered and novel stimuli as either remembered by a subject, or perceived as novel by a subject. However, these experiments have frequently been limited to memory for stimuli encountered in the lab shortly before the memory test. This study uses images collected at three time intervals (6 months, 3 months, and 2 weeks) using a wearable camera that regularly takes still photographs to investigate the performance of similar classifiers on memories for events that happen outside the lab, at multiple time intervals before the memory test.

Unsupervised Disambiguation of Preposition Senses with an LDA Topic Model (Miriam Connor, advised by Beth Levin, Linguistics)
Though it has received relatively little attention in the sense disambiguation literature, preposition sense disambiguation (PSD) represents a challenging task with important applications in machine translation and relation extraction. Most work on PSD has involved on supervised systems, but only a small amount of reliable annotated data is available for preposition sense. I present an unsupervised model for PSD, which performs sense discrimination using a Latent Dirichlet Allocation model and discovers semantic relations among the prepositions with group average agglomerative clustering. I compare my system’s performance with previous work on both supervised and unsupervised PSD models and suggest future directions and applications for PSD.

Peter Jenks (Berkeley) will be visiting on Tuesday for the SMircle meeting at 2pm in the Greenberg Room. He will be talking about his research on quantifier scope. Come join!

Quantifier Float and Scope in Thai

In this paper I present an analysis of quantifier float (Q-float) in Thai as overt Quantifier Raising (QR). I show that floated quantifiers have at least four properties which are typically associated with QR: locally restricted to arguments, rigid scope, and an adverbial distribution. I argue Thai Q-float cannot be generated by stranding (Sportiche 1988, Miyagawa 1989) or extraposition (Simpson 2009). I propose instead that Thai Q-float is QR, here conceived as A-scrambling to a projection of VP (Johnson and Tomioka 1997). I further show that Thai Q-float provides evidence for the syntactic mechanisms that underlie QR: quantifiers in Thai form a natural class with adverbs and verbs in their ability to be negated and their ability to form the elliptical answer to a polar question. This indicates that quantifiers are syntactically both adnominal and adverbial, an observation which matches the traditional view of Generalized Quantifiers (Barwise and Cooper 1981). Finally, I show that Q-float in Thai is driven by information focus on the quantifier, and that Thai QR is otherwise covert.

Who’s talking around these days?

• At Princeton: Chris Potts this week and Dan Jurafsky next week.

Come to the Greenberg Room this afternoon from 2-5pm for talks by grad students Hsin-Chang Chen, James Collins, Judy Kroo, Dasha Popova, and Tania Rojas-Esponda. Special bonus: Social to follow hosted by the QP Fest Committee – so our first-years can just relax and enjoy the talks! Visit this link for agenda and abstracts (updated as more become available).

Come to the Greenberg Room on Monday 4/22 from 12:15-1:05 for a SymSys Forum. Jeff Shrager will be giving a talk entitled “Aspects of the Development of Commonsense Perception”.

Commonsense Perception (sometimes called View Application or Conceptual Blending) is among the most important cognitive skills. Jeff Shrager will be speaking about neuro-psychological, developmental, and computational approaches to understanding how Commonsense Perception works, and how children learn to do it.