AQFC2015

Towards embeddings for information retrieval

----------------------------------------------------------------------------------------------------------------------------

Department of Systems Engineering and Engineering Management
The Chinese University of Hong Kong

----------------------------------------------------------------------------------------------------------------------------

Date: 4:30pm-5:30pm, December 9 (Friday), 2016

Title: Towards embeddings for information retrieval

Speaker: Dr. Shoaib Jameel, Cardiff University

Abstract:
We present two recent models to generate embeddings. In the first model, we concentrate on generating conceptual subspaces. Conceptual spaces are geometric representations of conceptual knowledge in which entities correspond to points, natural properties correspond to convex regions, and the dimensions of the space correspond to salient features. While conceptual spaces enable elegant models of various cognitive phenomena, the lack of automated methods for constructing such representations have so far limited their application in artificial intelligence. To address this issue, we propose a method which learns a vector-space embedding of entities from Wikipedia and constrains this embedding such that entities of the same semantic type are located in some lower-dimensional subspace. In the second part of the talk, we also propose a new word embedding model, inspired by GloVe, which is formulated as a feasible least squares optimization problem. In contrast to existing models, we explicitly represent the uncertainty about the exact definition of each word vector. To this end, we estimate the error that results from using noisy co-occurrence counts in the formulation of the model, and we model the imprecision that results from including uninformative context words. Our experimental results demonstrate that these models compare favourably with existing models.

Biography:
Shoaib Jameel is currently a postdoctoral researcher at the School of Computer Science and Informatics, Cardiff University in the United Kingdom, working on an ERC project with Dr. Steven Schockaert. He has obtained his PhD from the Department of Systems Engineering and Engineering Management (SEEM), The Chinese University of Hong Kong in the year 2014 under the supervision of Prof. Wai Lam. His research interests include probabilistic topic models, word embedding models and information retrieval.

Everyone is welcome to attend the talk!

Venue: Room 513,
William M.W. Mong Engineering Building (ERB),
(Engineering Building Complex Phase 2)
The Chinese University of Hong Kong.

The talk will be hosted by:
Prof. Wai Lam,
Department of Systems Engineering and Engineering Management,
The Chinese University of Hong Kong,
E-mail: wlam@se.cuhk.edu.hk
Homepage: http://www1.se.cuhk.edu.hk/~textmine/

SEEM-5201 Website: http://seminar.se.cuhk.edu.hk
Email: seem5201@se.cuhk.edu.hk

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

Date: 
Friday, December 9, 2016 - 08:30 to 09:30