Pitfalls of studying language in isolation

Studies of language acquisition and language understanding display a remarkable lack of attention to the subject matter of the utterances being studied.  This is probably because nobody knows how to represent and process meaning whereas the forms of utterances are readily available.  Thus "language acquisition" have come to mean the study of learning how to construct utterances "of the right form" and studies of language understanding focus on translating forms of utterances into other symbolic forms equally devoid of the richness and detail of the things the utterance is supposed to convey.

A real theory of language acquisition should study how babies learn to decode form-meaning mappings in an environment where lots of things are going on in addition to what is being said.  A real theory of language understanding should study what kinds of rich interconnected concepts and embodied simulations get triggered by words and constructions, how we decide what to simulate given the scant detail in descriptions, and what inferences are made possible beyond what is explicitly stated.  

All this is AI-complete you say?  Well by limiting ourselves to study language in isolation, we may have come to the end of the line where the ~80% accuracy limit of machine learning based computational linguistics (on almost any linguistic problem you can think of) is preventing us from building truly transformative applications.  Maybe we are shooting ourselves in the foot, and maybe, just maybe, some problems that look difficult right now are difficult not because we are missing the right machine learning algorithm or sufficient labeled data but because we are ignoring the constraints imposed by the meaning side of things.  We may have finally run out of options other than to try and crack the real problem, i.e. modeling what utterances are ABOUT.

Full post...

Bret Victor

Bret Victor - Inventing on Principle from CUSEC on Vimeo.

Bret Victor's inspirational talk with his views on (1) how to flourish fragile ideas, and (2) how to live your life. For more from Bret, check out his website.
Full post...

Vi Hart

For more from my 7 year old daughter's new favorite educator Vi Hart check out Khan Academy, YouTube, other videos, Wikipedia, or her blog.
Full post...

A Mathematician's Lament by Paul Lockhart

This is a wonderful and heartfelt book about the poetry of mathematics and how we fail to introduce it to students at schools and to society in general. I can finally stop feeling guilty about my obsession with popular math books and cool problems even though I know they probably have no practical value to me or to society whatsoever. After all it is not practical value that makes reading poetry a pleasure! I am thankful to all my wonderful math teachers who bequeathed me this guilty pleasure without my knowledge or consent in spite of the horrible curriculum, text books, and test system. And I highly recommend the author's second book "Measurement" to those who seek an alternative to the typical 10 pound high school textbooks full of repetitive and unimaginative problems.

Full post... Related link

Parser Evaluation Using Textual Entailments

Deniz Yuret, Laura Rimell, Aydin Han. Language Resources and Evaluation. 2012. (PDF, URL, Task website, Related posts)

Abstract
Parser Evaluation using Textual Entailments (PETE) is a shared task in
the SemEval-2010 Evaluation Exercises on Semantic Evaluation. The
task involves recognizing textual entailments based on syntactic
information alone. PETE introduces a new parser evaluation scheme
that is formalism independent, less prone to annotation error, and
focused on semantically relevant distinctions. This paper describes
the PETE task, gives an error analysis of the top-performing Cambridge
system, and introduces a standard entailment module that can be used
with any parser that outputs Stanford typed dependencies.

Full post...