Doyen Sahoo

Doyen Sahoo

Research Scientist
Living Analytics Research Centre
School of Information Systems
​Singapore Management University

About Me

Picture
I am Doyen, a Research Scientist at Living Analytics Research Centre (LARC), Singapore Management University (SMU). I am also serving as an Adjunct Faculty in SMU's MITB Programme in the Artificial Intelligence Track.  

I have been actively pursuing machine learning research for over six years, which has resulted in several publications at top tier venues. Currently, I am working on  Smart Consumption and Healthy Lifestyle projects at LARC, contributing to research and development in the FoodAI project. I am also working on Deep Learning Research, investigating algorithms for learning with deep neural networks for various problem settings. 

In the past, my research interests have spanned both fundamental and applied machine learning. I have worked on investigating novel methods for Online Learning, Kernel Methods, and Deep Learning. I have also worked on applying machine learning techniques for computer vision applications, computational finance and cyber security. 

I obtained my Ph.D. in Information Systems from Singapore Management University under the guidance of Professor Steven Hoi, where my thesis focused on development of novel online learning algorithms. I obtained my B.Eng. in Computer Science from Nanyang Technological University.

Research Projects

These are some of the major research projects I have worked on in recent years
Picture
FoodAI uses state of the art visual recognition technology, based on Deep Learning, to enable smart food recognition. It is designed to recognize over 750 commonly found food categories in Singapore. ​This technology is currently in use by several organisations globally. 
[
Demo]

Picture
AntiMalweb investigates machine learning techniques for Malicious URL Detection. Traditional approaches rely on blacklists, which can not be exhaustive, and are not useful for newly generated URLs. Using machine learning solutions, thus, offers a promising direction to solve this problem.
[Demo][GitHub][Malicious URL Detection Survey]

Picture
LIBOL is a Library of Online Learning algorithms, a class of machine learning algorithms that learn from a stream of data where instances arrive sequentially.  The on-the-fly nature of these algorithms makes them extremely scalable.
[GitHub][Toolbox - JMLR][Online Learning Survey]

Picture
Online Portfolio Selection deals with application of online learning to determine the wealth allocation into a portfolio of assets with the goal to optimize a certain metric (e.g. cumulative wealth, risk-adjusted return, etc.)
[GitHub][Toolbox - YouTube][Toolbox - JMLR][OLPS Survey]

​Publications [Google Scholar]

Preprints

Online Learning: A Comprehensive Survey
​
Steven C. H. Hoi, Doyen Sahoo, Jing Lu, Peilin Zhao
​
arXiv preprint arXiv:1802.02871 (2018).
​[arXiv]
Meta-Learning with Domain Adaptation for Few-Shot Learning under Domain Shift
 Doyen Sahoo, Hung Le, Chenghao Liu
, Steven C. H. Hoi,
​[pdf]
URLNet: Learning a URL Representation with Deep Learning for Malicious URL Detection
​Hung Le, Quang Pham, Doyen Sahoo, Steven C. H. Hoi
arXiv preprint arXiv:1802.03162 (2018)
[arXiv] [MIT Tech Review] [Jack-Clark]
Malicious URL detection using machine learning: A survey
​
Doyen Sahoo,
Chenghao Liu, Steven C. H. Hoi
arXiv preprint arXiv:1701.07179 (2017)
[arXiv] 
Recent Advances in Deep Learning for Object Detection
​Xiongwei Wu, Doyen Sahoo, Steven C. H. Hoi
arXiv preprint arXiv:1908.03673 (2019)
[arXiv] 


2019

Multimodal Transformer Networks for End-to-End Video-Grounded Dialogue System
​Hung Le, Doyen Sahoo, Nancy Chen, Steven C. H. Hoi
Association for Computational Linguistics Conference (ACL) 2019
FoodAI: Food Image Recognition via Deep Learning for Smart Food Logging
​Doyen Sahoo, 
Wang Hao, Shu Ke, Xiongwei Wu, Hung Le, Palakorn Achananuparp, Ee-Peng Lim, Steven C. H. Hoi
Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD) 2019
Learning Cross-Modal Embeddings with Adversarial Networks for Cooking Recipes and Food Images​
​Wang Hao*, Doyen Sahoo*,  Chenghao Liu, Ee-Peng Lim, Steven C. H. Hoi
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2019
​* Equal Contribution

[arXiv] [GitHub]
Large Scale Online Multiple Kernel Regression with Application to Time-Series Prediction
​Doyen Sahoo,
 Steven C. H. Hoi, Bin Li

ACM Transactions on Knowledge Discovery from Data (TKDD) 2019
End-to-End Multimodal Dialog Systems with Hierarchical Multimodal Attention of Video Features
Hung Le, Steven C.H. Hoi, Doyen Sahoo, Nancy F. Chen​
7th Dialogue System Technology Challenge (DSTC7) Workshop, (AAAI), 2019

2018

Online Deep Learning: Learning Deep Neural Networks on the fly​
Doyen Sahoo, Quang Pham, Jing Lu, Steven C. H. Hoi

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI) 2018
[pdf][arXiv][GitHub]
Sparse Passive-Aggressive learning for bounded online kernel methods
Jing Lu, Doyen Sahoo, Peilin Zhao, Steven C. H. Hoi
ACM Transactions on Intelligent Systems and Technology (TIST) 2018
[pdf]
SOL: A library for scalable online learning algorithms
Yue Wu, Steven C. H. Hoi, Chenghao Liu, Jing Lu, Doyen Sahoo, Nenghai Yu
Neurocomputing 2018
[pdf]

2017 and before

Cost sensitive online multiple kernel classification
Doyen Sahoo, Peilin Zhao, Steven C. H. Hoi
Asian Conference on Machine Learning (ACML) 2016
[pdf]
Temporal kernel descriptors for learning with time-sensitive patterns
Doyen Sahoo,  Abhishek Sharma, Steven C. H. Hoi, Peilin Zhao
Proceedings of the 2016 SIAM International Conference on Data Mining (SDM) 2016
[pdf]
OLPS: A toolbox for on-line portfolio selection
Bin Li, Doyen Sahoo, Steven C. H. Hoi
The Journal of Machine Learning Research (JMLR) 2016
[pdf]
Moving average reversion strategy for on-line portfolio selection
Bin Li, Steven C. H. Hoi, Doyen Sahoo, Zhi-Yong Liu
Artificial Intelligence  2015
[pdf]
Online multiple kernel regression
Doyen Sahoo, Steven C. H. Hoi, Bin Li
ACM SIGKDD international conference on Knowledge discovery and data mining (KDD) 2014
[pdf]
Powered by Create your own unique website with customizable templates.