Research Interests:

In general, I am interested in Machine Learning and its applications in real world problems. I am currently interested in semi-supervised and active learning methods for activity recognition, human behaviour understanding and user modeling. In the past couple of years I was involved in the following areas:

1) Gaussian Random Fields

2) Krylov Subspace Methods

3) Multi-pole methods and N-body Learning

4) Document Classification

5) Object Class Recognition

 

Refereed Publications:

Maryam Mahdaviani, Nando de Freitas, Bob Fraser and Firas Hamze. Fast Computational Methods for Visually Guided Robots. ICRA 2005(Acceptance rate = 43%). [PDF]

Nando de Freitas, Yang Wang, Maryam Mahdaviani, Dustin Lang. Fast Krylov Methods for N-Body Learning. NIPS 2005(Acceptance rate = 27%). [PDF]

Maryam Mahdaviani and Nando de Freitas . User Adaptive Image Ranking for Search Engines. User Adaptive Systems workshop (acceptance rate = 33%), NIPS 2006.

Maryam Mahdaviani and Tanzeem Choudhury . Fast and Scalable Training of Semi-Supervised CRFs with Application to Activity Recognition . NIPS 2007 (Acceptance rate = 22%).

Miresmailli, S., Badulescu, D., Mahdaviani, M., Zamar, R. H. and Isman, M.B. 2009. Integrating plant chemical ecology, sensors and artificial intelligence for accurate pest monitoring. In: Columbus F. (Ed). Tomatoes: Agricultural procedures, pathogen interaction and health effects. Nova Science Publishers, NY. In Press

 

Other Publications:

My masters thesis: Semi-supervised and Active Trainng of Conditional Random Fields for Activity Recognition. [PDF]

My undergrad honours thesis: Fast Object Class Recognition. [PDF]

Tech Report: Integrating Gaussian Processes with Word-Sequence Kernels for Bayesian Text Categorization, with Sara Forghanizadeh and Giuseppe Carenini[ps.gz]

An updated version of this technical report has been submitted to HLT-NAACL 2007.

 

Presentations:

ICRA 2005 [PPT]

NIPS 2005 [PPT]

NIPS 2006 User Adaptive Systems Workshop [PPT]

 

Last update: August 14, 2009