Posts by Collection

portfolio

publications

MOGT: oversampling with a parsimonious mixture of Gaussian trees model for imbalanced time-series classification

Published in IEEE International Workshop on Machine Learning for Signal Processing (MLSP), 2013

This paper is on using a mixture of gaussian trees statistical model to solve the problem of multi-modal imbalanced time-series classification.

Recommended citation: John Z. F. Pang, Hong Cao, Vincent Y. F. Tan (2013). "MOGT: oversampling with a parsimonious mixture of Gaussian trees model for imbalanced time-series classification." IEEE Machine Learning for Signal Processing. http://j-pang.github.io/files/mlsp2013_mogt.pdf

A parsimonious mixture of Gaussian trees model for oversampling in imbalanced and multimodal time-series classification

Published in IEEE Transactions of Neural Networks and Learning Systems, 2014

This paper extends the previous mixture of gaussian trees framework, including additional analysis and extensive experiments.

Recommended citation: Hong Cao, Vincent Y. F. Tan, John Z. F. Pang (2014). "A parsimonious mixture of Gaussian trees model for oversampling in imbalanced and multimodal time-series classification." IEEE Transactions of Neural Networks and Learning Systems. http://j-pang.github.io/files/paper2.pdf

Asymmetric optimal-velocity car-following model

Published in Physica A: Statistical Mechanics and its Applications, 2015

An asymmetric optimal velocity model is proposed for car-following theory, where the asymmetry between the acceleration and the deceleration is represented by the exponential function with an asymmetrical factor, which agrees with published experiments.

Recommended citation: Xihua Xu, John Z. F. Pang, Christopher Monterola (2015). "Asymmetric optimal-velocity car-following model." Physica A: Statistical Mechanics and its Applications. http://j-pang.github.io/files/phyA2015_aov.pdf

Cluster statistics and quasisoliton dynamics in microscopic optimal-velocity models

Published in Physical Review E, 2016

In this paper, we show that there exists an emergent intrinsic scale that characterizes the interaction strength between multiple clusters appearing in the solutions of optimal-velocity models. The interaction characterizes the dynamics of the localized quasisoliton structures given by the time derivative of the headways, and the intrinsic scale is analogous to the “charge” of the quasisolitons, leading to non-trivial cluster statistics from the random perturbations to the initial steady states of uniform headways. The cluster statistics depend both on the quasisoliton charge and the density of the traffic.

Recommended citation: Bo Yang, Xihua Xu, John Z. F. Pang, Christopher Monterola (2016). "Cluster statistics and quasisoliton dynamics in microscopic optimal-velocity models." Physical Review E. http://j-pang.github.io/files/prE2016_cluster.pdf

Dynamic Service Function Chaining in SDN-Enabled Networks with Middleboxes

Published in IEEE International Conference on Network Protocols (ICNP), 2016

This paper utilizes a primal-dual update algorithm to optimize dynamic service function chaining in SDN-enabled networks.

Recommended citation: Linqi Guo, John Z. F. Pang, Anwar Walid (2016). "Dynamic Service Function Chaining in SDN-Enabled Networks with Middleboxes." IEEE International Conference on Network Protocols. http://j-pang.github.io/files/icnp2016_dsfc.pdf

The efficiency of open access in platforms for networked cournot markets

Published in IEEE Conference on Computer Communications (INFOCOM), 2017

This paper studies how the efficiency of an online platform is impacted by the degree to which access of platform participants is open or controlled.

Recommended citation: John Z. F. Pang, Hu Fu, Won I. Lee, Adam Wierman (2017). "The efficiency of open access in platforms for networked cournot markets." IEEE Conference on Computer Communications (INFOCOM). http://j-pang.github.io/files/infocom2017_netcournot.pdf

Optimal load control for frequency regulation under limited control coverage

Published in IREP Symposium, 2017

By framing the problem as an optimization problem and applying saddle-point dynamics, we obtain a control law that rebalances power and asymptotically stabilizes frequency after a disturbance. In this paper, we generalize previous work to partial control over all nodes in the network.

Recommended citation: John Z. F. Pang, Linqi Guo, Steven H Low (2017). "Optimal load control for frequency regulation under limited control coverage." IREP Symposium. http://j-pang.github.io/files/irep2017_optloadcontrol.pdf

Networked cournot competition in platform markets: Access control and efficiency loss

Published in IEEE Annual Conference for Decision and Control (CDC), 2017

This paper extends the 2017 INFOCOM paper through a key lemma that proves a worst case cost function in computing Price of Anarchy results.

Recommended citation: Weixuan Lin, John Z. F. Pang, Eilyan Bitar, Adam Wierman (2017). "Networked cournot competition in platform markets: Access control and efficiency loss." IEEE Annual Conference for Decision and Control (CDC). http://j-pang.github.io/files/cdc2017_netcournot.pdf

Battery swapping assignment for electric vehicles: A bipartite matching approach

Published in IEEE Annual Conference on Decision and Control (CDC), 2017

This paper models the battery swapping assignment problem as a bipartite matching problem, utilizing the Hungarian Algorithm, which provides significant improvement over greedy algorithms.

Recommended citation: Pengcheng You, John Z. F. Pang, Minghua Chen, Steven H Low, Youxian Sun (2017). "Battery swapping assignment for electric vehicles: A bipartite matching approach." IEEE Annual Conference on Decision and Control (CDC). http://j-pang.github.io/files/cdc2017_offlinebattswap.pdf

Deep Koopman Controller Synthesis for Cyber-Resilient Market-Based Frequency Regulation

Published in IFAC Symposium on Control of Power and Energy Systems (CPES), 2018

This paper proposes a deep Koopman controlled approach for market-based frequency regulation.

Recommended citation: Pengcheng You, John Z. F. Pang, Enoch Yeung (2018). "Deep Koopman Controller Synthesis for Cyber-Resilient Market-Based Frequency Regulation." IFAC Symposium on Control of Power and Energy Systems (CPES). http://j-pang.github.io/files/ifac2018_koopmancontrol.pdf

Joint placement and routing of network function chains in data centers

Published in IEEE Conference on Computer Communications (INFOCOM), 2018

In this work, we investigate the problem of joint optimization over placement and routing of network function chains in data centers.

Recommended citation: Linqi Guo, John Z. F. Pang, Anwar Walid (2018). "Joint placement and routing of network function chains in data centers." IEEE Conference on Computer Communications (INFOCOM). http://j-pang.github.io/files/infocom2018_prnfc.pdf

Efficient Online Station Assignment for EV Battery Swapping

Published in Ninth International Conference on Future Energy Systems (e-Energy), 2018

This paper extends the offline bipartite matching solution and investigates the online station assignment for (commercial) electric vehicles (EVs) that make battery swapping requests to a central operator, with the aim of minimizing cost to EVs and congestion at service stations.

Recommended citation: Pengcheng You, Peng Cheng, John Z. F. Pang, Steven H Low (2018). "Efficient Online Station Assignment for EV Battery Swapping." Ninth International Conference on Future Energy Systems (e-Energy). http://j-pang.github.io/files/eenergy2018_onlinebattswap.pdf

Distributed optimal frequency control considering a nonlinear network-preserving model

Published in IEEE Transactions on Power Systems, 2018

This paper extends the IREP symposium paper, addressing the distributed optimal frequency control of power systems considering a network-preserving model with nonlinear power flows and excitation voltage dynamics.

Recommended citation: Zhaojian Wang, Feng Liu, John Z. F. Pang, Steven H Low, Shengwei Mei (2018). "Distributed optimal frequency control considering a nonlinear network-preserving model." IEEE Transactions on Power Systems. http://j-pang.github.io/files/tps2018_optloadcontrol.pdf

Transparency and Control in Platforms for Networked Markets

Published in In submission to Operations Research, 2019

This paper extends the INFOCOM 2017 and CDC 2017 paper and analyze the worst case efficiency loss of online platform designs under a networked Cournot competition model. The manuscript is currently in final revision.

Recommended citation: John Z. F. Pang, Weixuan Lin, Hu Fu, Jack Kleeman, Eilyan Bitar, Adam Wierman (2019). "Transparency and Control in Platforms for Networked Markets." In submission to Operations Research. http://j-pang.github.io/files/arxiv2019_netcournot.pdf

Competitive online optimization under inventory constraints

Published in ACM Conference on Measurement and Analysis of Computing Systems (SIGMETRICS), 2019

This paper studies online optimization under inventory (budget) constraints.

Recommended citation: Qiulin Lin, Hanling Yi, John Z. F. Pang, Minghua Chen, Adam Wierman, Michael Honig, Yuanzhang Xiao (2019). "Competitive online optimization under inventory constraints." ACM Conference on Measurement and Analysis of Computing Systems (SIGMETRICS). http://j-pang.github.io/files/sigmetrics2019_onlineopti.pdf

talks

teaching