Sarah Erfani
Lecturer (equiv US Assistant Professor)
Department of Computing and Information Systems
Melbourne School of Engineering
The University of Melbourne
Contact
Email: [firstname].[lastname]@unimelb.edu.au
Location: Level 07, Room 7.16,
Doug McDonell Building, The University of Melbourne,
Parkville, VIC 3010, Australia
Phone: 03 90358156
Research Interests
- Machine Learning
- Artificial Intelligence
- Large-Scale Data Mining
- Time Series Analysis
- Cyber Security
- Data Privacy
Publications
2019
• Fateme Fahiman, Steven Disano, Sarah Erfani, Pierluigi Mancarella, Christopher Leckie, “Data Driven Dynamic Probabilistic Reserve Sizing Based on Dynamic Bayesian Belief Networks”, IEEE Transactions on Power Systems, 2019.
2018
- Shuo Zhou, Sarah Erfani, James Bailey, “Online CP Decomposition for Sparse Tensors”, IEEE International Conference on Data Mining (ICDM), 2018. pdf
- Masud Moshtaghi, James Bezdek, Sarah M. Erfani, Christopher Leckie, James Bailey, “Online Cluster Validity Indices for Streaming Data”, International Journal of Intelligent Systems (IJIS), 2018. pdf
- Yi Han, Benjamin Rubinstein, Tamas Abraham, Tansu Alpcan, Olivier De Vel, Sarah Erfani, David Hubczenko, Christopher Leckie and Paul Montague, “Reinforcement Learning for Autonomous Defence in Software-Defined Networking”, International Conference on Decision and Game Theory for Security (GameSec), 2018. pdf
- Sandamal Weerasinghe, Tansu Alpcan, Sarah M. Erfani, Christopher Leckie, Peyam Pourbeik and Jack Riddle, “Deep Learning Based Game-Theoretical Approach to Evade Jamming Attacks”, International Conference on Decision and Game Theory for Security (GameSec), 2018. pdf
- Yisen Wang, Bo Dai, Lingkai Kong, Hongyuan Zha, Xingjun Ma, Sarah M. Erfani, James Bailey, Le Song, Shu-Tao Xia, “Learning Deep Hidden Nonlinear Dynamics from Aggregate Data”, Conference on Uncertainty in Artificial Intelligence (UAI), 2018. pdf
- Xingjun Ma, Yisen Wang, Michael E. Houle, Shuo Zhou, Sarah M. Erfani, Shu-Tao Xia, Sudanthi Wijewickrema, James Bailey. “Dimensionality-Driven Learning with Noisy Labels”, International Conference on Machine Learning (ICML), p. 10, 2018. pdf
- Weihao Cheng, Sarah M Erfani, Rui Zhang, Kotagiri Ramamohanarao,“Predicting Complex Activities from Ongoing Multivariate Time Series”, International Joint Conference on Artificial Intelligence (IJCAI), p. 7, 2018. pdf
- Prameesha Sandamal Weerasinghe, Sarah Erfani, Tansu Alpcan, Christopher Leckie, “Detection of Anomalous Communications with SDRs and Unsupervised Adversarial Learning”, IEEE Conference on Local Computer Networks (LCN), p. 4, 2018.
- Prameesha Sandamal Weerasinghe, Tansu Alpcan, Sarah Erfani, Christopher Leckie, “Unsupervised Adversarial Anomaly Detection using One-Class Support Vector Machines”, International Symposium on Mathematic Theory of Networks and Systems (MTNS), p. 4, 2018. pdf
- Xingjun Ma, Bo Li, Yisen Wang, Sarah Erfani, Sudanthi Wijewickrema, Michael Houle, Grant Schoenebeck, Dawn Song, James Bailey, “Charactrizing Adversarial Subspaces Using Local Intrinsic Dimensionality”, International Conference on Learning Representations (ICLR), p.15, 2018. pdf Accepted in top 2%
- Zahra Ghafoori, Sarah Erfani, Sutharshan Rajasegarar, James Bezdek, Shanika Karunasekera, Christopher Leckie, “Efficient Unsupervised Parameter Estimation for One-Class Support Vector Machines”, IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2018. pdf
- Weihao Cheng, Sarah M Erfani, Rui Zhang, Kotagiri Ramamohanarao,“Learning Datum-Wise Sampling Frequency for Energy-Efficient Human Activity Recognition”, Association for the Advancement of Artificial Intelligence (AAAI), p. 8, 2018. pdf
2017
- Shuo Zhou, Sarah Erfani, James Bailey, “SCED: A General Framework for Sparse Tensor Decompositions with Constraints and Elementwise Dynamic Learning”, IEEE International Conference on Data Mining (ICDM),p. 10, 2017. pdf
- Li Li, Sarah Erfani, and Christopher Andrew Leckie, “A Pattern Tree based Method for Mining Conditional Contrast Pattern from Multi-Source Data”, IEEE International Conference on Data Mining Workshops (ICDMW), p. 8, 2017. pdf
- Laurent Amsaleg, James Bailey, Sarah Erfani, Teddy Furon, Michael E Houle, Miloš Radovanovic, Nguyen Xuan Vinh, “The Vulnerability of Learning to Adversarial Perturbation Increases with Intrinsic Dimensionality”, IEEE Workshop on Information Forensics and Security (WIFS), 2017
- Elaheh Alipourchavary, Sarah M. Erfani, Christopher Leckie, “Summarizing Significant Changes in Network Traffic Using Contrast Pattern Mining”, ACM International Conference on Information and Knowledge Management (CIKM), p. 4, 2017. pdf
- Weihao Cheng, Sarah M. Erfani, Ruiand Zhang, Kotagiri Ramamohanarao, “Markov Dynamic Subsequence Ensemble for Energy-Efficient Activity Recognition”, International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services (MobiQuitous), p. 10, 2017. pdf
- Punit Rathore, James C. Bezdek, Sarah M. Erfani, Sutharshan Rajasegarar, Marimuthu Palaniswami, “Ensemble fuzzy clustering using random projections of high dimensional data”, IEEE Transactions on Fuzzy Systems (TFS), p. 15, 2017. pdf
- Weihao Cheng, Sarah M Erfani, Rui Zhang, Kotagiri Ramamohanarao,“Accurate Recognition of the Current Activity in the Presence of Multiple Activities”, Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), p. 12, 2017. pdf
- Timothy B. Iredale, Sarah M. Erfani, Christopher Leckie, “An efficient visual assessment of cluster tendency tool for large-scale time series data sets”, IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), p. 8, 2017. pdf
- Fateme Fahiman, James C. Bezdek, Sarah M. Erfani, Marimuthu Palaniswami, Christopher Leckie, “Fuzzy c-Shape: A New Algorithm for Clustering Finite Time Series Waveforms”, IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), p. 8, 2017. pdf (Best paper award)
- Fateme Fahiman, Sarah M.Erfani, Sutharshan Rajasegarar, Marimuthu Palaniswami, Christopher Leckie, “Improving Load Forecasting Based on Deep Learning and K-shape Clustering”, International Joint Conference on Neural Networks (IJCNN), p. 8, 2017. pdf
- Sarah M. Erfani, Mahsa Baktashmotlagh, Masud Moshtaghi, Vinh Nguyen, Christopher Leckie, James Bailey, Kotagiri Ramamohanarao, “From shared subspaces to shared landmarks: a robust multi-source classification approach”, Association for the Advancement of Artificial Intelligence (AAAI), p. 7, 2017. pdf
- Masud Moshtaghi, Sarah M. Erfani, Christopher Leckie, James C. Bezdek, “Exponentially weighted ellipsoidal model for anomaly detection”, International Journal of Intelligent Systems (IJIS), 2017.
2016
- Laurent Amsaleg, James Bailey, Sarah Erfani, Teddy Furon, Michael E Houle, Miloš Radovanovic, Nguyen Xuan Vinh, “The vulnerability of learning to adversarial perturbation increases with intrinsic dimensionality”, NII Technical Report, p. 16, 2016. pdf
- Sarah M. Erfani, Mahsa Baktashmotlagh, Masud Moshtaghi, Vinh Nguyen, Christopher Leckie, James Bailey, Kotagiri Ramamohanarao, “Efficient domain generalisation using featurised kernel embeding”, International Joint Conference on Artificial Intelligence (IJCAI), p. 7, 2016. pdf
- Vinh Xuan Nguyen, Sarah Erfani, Sakrapee Paisitkriangkrai, James Bailey, Christopher Leckie, Kotagiri Ramamohanarao, “Training robust models using random projection”, International Conference on Pattern Recognition (ICPR), p. 6, 2016. pdf
- Sarah M. Erfani, Sutharshan Rajasegarar, Shanika Karunasekera, Christopher Leckie, “One-class support vector machine based deep learning for anomaly detection”, Pattern Recognition, p. 14, 2016. pdf (Best paper award)
- Zahra Ghafoori, Sarah M. Erfani,Sutharshan Rajasegarar, Shanika Karunasekera, Christopher Leckie, “Anomaly detection in non-stationary data: ensemble based self-adaptive OCSVM”, International Joint Conference on Neural Networks (IJCNN), p. 8, 2016. pdf
- Sarah M. Erfani, Mahsa Baktashmotlagh, Sutharshan Rajasegarar, Vinh Nguyen, Christopher Leckie, James Bailey, Kotagiri Ramamohanarao, “R1STM: One-class support tensor machine with randomised kernel”, SIAM International Conference in Data Mining (SDM), p. 9, 2016. pdf
- Timothy Glennan, Christopher Leckie, Sarah M. Erfani, "Improved classification of known and unknown network traffic flows using semi-supervised machine learning”, Australasian Conference on Information Security and Privacy (ACISP), p.8, 2016. pdf
- Zahra Ghafoori, Sutharshan Rajasegarar, Sarah M. Erfani, Shanika Karunasekera, Christopher Leckie, “Unsupervised parameter estimation for one-class support vector machines has been updated”, Pacific Asia Knowledge Discovery and Data Mining (PAKDD), p. 12, 2016. pdf
- Lingjuan Lyu, Yee Wei Law, Sarah M. Erfani, Christopher Leckie, Marimuthu Palaniswami, “An improved scheme for privacy-preserving collaborative anomaly detection”, IEEE PERCOM workshop on Security Privacy and Trust in IoT (SPT-IoT), p.6, 2016. pdf
Pre 2016
- Sarah M. Erfani, Mahsa Baktashmotlagh, Sutharshan Rajasegarar, Shanika Karunasekera, Christopher Leckie, "R1SVM: a randomised nonlinear approach to large-scale anomaly detection", Association for the Advancement of Artificial Intelligence (AAAI), 2015. pdf
- Sarah M. Erfani, Yee Wei Law, Shanika Karunasekera, Christopher Leckie, Marimuthu Palaniswami, "Privacy-preserving collaborative anomaly detection for participatory sensing", Pacific Asia Knowledge Discovery and Data Mining (PAKDD), 2014, pp. 581-593. pdf
- Sarah M. Erfani, Shanika Karunasekera, Cristopher Leckie, Udaya Parampalli, "Privacy-preserving data aggregation in participatory sensing networks", International Conference on Intelligent Sensors, Sensor Networks and Information Processing (ISSNIP), 2013, pp. 165-170.pdf
- Sarah M. Erfani, Sutharshan Rajasegarar, Christopher Leckie. "An efficient approach to detecting concept-evolution in network data streams." Australasian Telecommunication Networks and Applications Conference (ATNAC), 2011, pp. 1-7.pdf
Preprints
- Tansu Alpcan, Sarah Erfani, Christopher Leckie, “Toward the Starting Line: A Systems Engineering Approach to Strong AI”, p. 11, 2017. pdf
Recent Research Fundings
- Aligning Consent with Privacy Promises of Data Sharing Platforms, Network Society Institute Seed Funding, 2018-2019, $28k.
- Detection of Infected Internet-of-Thing (IoT) Devices to Prevent Distributed Denial of Service (DDoS) Attacks, Oceania Cyber Security Centre (OCSC) Proof of Concept Grant, 2018, $78k.
- Adversarial Machine Learning for Cyber, Defence Science & Technology Group and Data61 CRP, 2017-2020, $1.05m.
- Real-time Anomaly Detection in Large Scale High Dimensional Datasets, Huawei Innovation Research Program OPEN (HIRP), 2017-2019, $200k.
- Cyber security for the Internet of Things (IoT), Early Career Research (ECR) Grant, 2017, $40k.
Current PhD Students
- Yixin Su
- Jiabo He
- Sukarna Barua
- Ahmad Asghariyan Rezayi
- Elaheh Alipourchavary
- Namrata Srivastava
- Prameesha Weerasinghe
- Li Li
- Weihao Cheng
- Florin Schimbinschi
- Fateme Fahiman
- Shuo Zhou
Teaching
Projects
- Adversarial Machine Learning (AML). here
- Smart City and Internet of Things (IoT).here
Awards and Distinctions
- Best Paper Award: Fateme Fahiman, James C. Bezdek, Sarah M. Erfani, Marimuthu Palaniswami, Christopher Leckie, “Fuzzy c-Shape: A New Algorithm for Clustering Finite Time Series Waveforms”, IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), p. 8, 2017
- Best Paper Award: Sarah M. Erfani, Sutharshan Rajasegarar, Shanika Karunasekera, Christopher Leckie, “One-class support vector machine based deep learning for anomaly detection”, Pattern Recognition, p. 14, 2016.
- First prize in InfraHack competition, Melbourne infrastructure hack for Australia, 2014.here