I can also be found on Google Scholar.

Patents

2020

Binbin Li, Maxim Budninskiy, Joshua Rangsikitpho, Jamie Auslander, Kenneth Hwang, Amanda Conrad, and Simon Asselin, “Synchronizing Survey Collection with Ad Distribution During An Ad Campaign,” Application No.: 16/821,815, Filing Date: Mar. 17, 2020. Assignee: Disney Enterprises, Inc.. [Patent page→]

Abstract:
A system includes a hardware processor and a memory storing software code for controlling distribution of ad surveys to target audiences. For each target audience, the hardware processor executes the software code to receive ad data describing an ad volume distributed to the target audience during a predetermined time interval of an ad campaign, and to receive survey data describing a survey volume distributed to the target audience and a survey response volume collected from the target audience during the predetermined time interval. The software code also determines a next survey sampling rate for the target audience based on the ad data and the survey data, such that the volume of survey responses collected from the target audience during matches an ad distribution volume for the target audience within a predetermined threshold during each predetermined time interval of the ad campaign.

2019

Binbin Li, Amanda Conrad, Simon Asselin, Jamie Auslander, Joshua Rangsikitpho, Zhenyu Shi, and Alexander Vondrak, “Automated Advertisement Selection Using a Trained Predictive Model,” Application No.: 16/669,397, Filing Date: Oct. 30, 2019. Assignee: Disney Enterprises, Inc.. [Patent page→]

Abstract:
An automated advertisement selection system includes a computing platform having a hardware processor and a system memory storing a software code including a trained predictive model and a scoring module. The hardware processor executes the software code to receive an advertising query, the advertising query including a multiple parameters describing a target consumer group, and to identify, using the trained predictive model, candidate advertisements for the target consumer group based on the multiple parameters. The hardware processor also executes the software code to determine, using the scoring module, desirability scores for each one of the plurality of candidate advertisements, each of the desirability scores corresponding to a likelihood of each respective one of the plurality of candidate advertisements enticing the target consumer group, and to select one of the plurality of candidate advertisements based on the desirability scores for distribution to the target consumer group.

2016

Ankur Gupta, Brian Lee Duke, Binbin Li, and Prathaban Mookiah, “Systems and Methods for Travel-Related Anomaly Detection,” Pub. No.: US 2016/0203490 A1, Pub. Date: Jul. 14, 2016. Assignee: SAS Institute. [Patent page→]

Abstract:
A fraud score for a transaction in connection with an account is computed from retrieved data to indicate a probability of the account being in a compromised condition. A travel score is computed, wherein the computed travel score indicates a likelihood that a user of the account is traveling from a user home location at the time of the received transaction. A self-similarity score may be computed if the computed fraud score is above a predetermined threshold to indicate similarity of the received transaction to other transactions of the account in the set of prior transactions. A suggested action is determined, based on a fraud decisioning operation (and optionally the self-similarity score) and a travel decisioning operation using the fraud score and travel score, respectively.

Publications

2012

Ioannis Ch. Paschalidis, Binbin Li, and Michael C. Caramanis, “Demand-Side Management for Regulation Service Provisioning Through Internal Pricing,” in IEEE Transactions on Power Systems, vol. 27, no. 3, pp. 1531-1539, Aug. 2012. [IEEE journal→]

Abstract:
We develop a market-based mechanism that enables a building smart microgrid operator (SMO) to offer regulation service reserves and meet the associated obligation of fast response to commands issued by the wholesale market independent system operator (ISO) who provides energy and purchases reserves. The proposed market-based mechanism allows the SMO to control the behavior of internal loads through price signals and to provide feedback to the ISO. A regulation service reserves quantity is transacted between the SMO and the ISO for a relatively long period of time (e.g., a one-hour-long time-scale). During this period the ISO follows shorter time-scale stochastic dynamics to repeatedly request from the SMO to decrease or increase its consumption. We model the operational task of selecting an optimal short time-scale dynamic pricing policy as a stochastic dynamic program that maximizes average SMO and ISO utility. We then formulate an associated nonlinear programming static problem that provides an upper bound on the optimal utility. We study an asymptotic regime in which this upper bound is shown to be tight and the static policy provides an efficient approximation of the dynamic pricing policy. Equally importantly, this framework allows us to optimize the long time-scale decision of determining the optimal regulation service reserve quantity. We demonstrate, verify and validate the proposed approach through a series of Monte Carlo simulations of the controlled system time trajectories.

2011

Ioannis Ch. Paschalidis and Binbin Li, “Energy Optimized Topologies for Distributed Averaging in Wireless Sensor Networks,” in IEEE Transactions on Automatic Control, vol. 56, no. 10, pp. 2290-2304, Oct. 2011. [IEEE journal→] [Poster→]

Abstract:
We study the energy efficient implementation of averaging/consensus algorithms in wireless sensor networks. For static, time-invariant topologies we start from the recent result that a bidirectional spanning tree is preferable in terms of convergence time. We formulate the combinatorial optimization problem of selecting such a minimal energy tree as a mixed integer linear programming problem. Since the problem is NP-complete we devise a semi-definite relaxation and establish bounds on the optimal cost. We also develop a series of graph-based algorithms that yield energy efficient bidirectional spanning trees and establish associated bounds on the optimal cost. For dynamic, time-varying topologies we consider a recently proposed load-balancing algorithm which has preferable convergence time properties. We formulate the problem of selecting a minimal energy interconnected network over which we can run the algorithm as a sequential decision problem and cast it into a dynamic programming framework. We first consider the scenario of a large enough time horizon and show that the problem is equivalent to constructing a Minimum Spanning Tree. We then consider the scenario of a limited time horizon and employ a “rollout” heuristic that leverages the spanning tree solution and yields efficient solutions for the original problem. Some of our algorithms can be run in a distributed manner and numerical results establish that they produce near-optimal solutions.

Ioannis Ch. Paschalidis, Binbin Li, and Michael C. Caramanis, “A market-based mechanism for providing demand-side regulation service reserves,” 2011 50th IEEE Conference on Decision and Control and European Control Conference, Orlando, FL, Dec. 2011, pp. 21-26. [IEEE paper→]

Abstract:
We develop a market-based mechanism that enables a building Smart Microgrid Operator (SMO) to offer regulation service reserves and meet the associated obligation of fast response to commands issued by the wholesale market Independent System Operator (ISO) who provides energy and purchases reserves. The proposed market-based mechanism allows the SMO to control the behavior of internal loads through price signals and to provide feedback to the ISO. A regulation service reserves quantity is transacted between the SMO and the ISO for a relatively long period of time (e.g., a one hour long time-scale). During this period the ISO repeatedly requests from the SMO to decrease or increase its consumption. We model the operational task of selecting an optimal short time-scale dynamic pricing policy as a stochastic dynamic program that maximizes average SMO and ISO utility. We then formulate an associated non-linear programming static problem that provides an upper bound on the optimal utility. We study an asymptotic regime in which this upper bound is tight and the static policy provides an efficient approximation of the dynamic pricing policy. We demonstrate, verify and validate the proposed approach through a series of Monte Carlo simulations of the controlled system time trajectories.

Ioannis Ch. Paschalidis and Binbin Li, “On Energy Optimized Network Construction for Distributed Averaging in a Dynamic Environment,” 2011 18th International Federation of Automatic Control (IFAC) World Congress, Milano, Italy, Aug. 2011, pp. 14958-14963. In: IFAC Proceedings Volumes, vol. 44, no. 1. [Elsevier paper→]

Abstract:
We study the energy costs of running a distributed averaging/consensus algorithm in a dynamic, time-varying wireless sensor network. It has been recently shown in Olshevsky and Tsitsiklis [2009] that running a load-balancing algorithm over a symmetric interconnected network is preferable in terms of convergence time. We formulate the problem of selecting a minimal energy interconnected network as a sequential decision problem and cast it into a Dynamic Programming (DP) framework. This problem is hard to solve, especially when incurring a penalty cost for not reaching interconnectivity within a pre-determined block of time. We first consider the scenario of a large enough time horizon and show that solving the DP is equivalent to constructing a Minimum Spanning Tree (MST), which can be done in a distributed manner. We then consider the scenario of a limited time horizon and employ a “rollout” heuristic that leverages the MST solution and yields efficient solutions for the original DP. Numerical experiments verify the effectiveness and efficiency of our algorithms.

2009

Ioannis Ch. Paschalidis and Binbin Li, “On energy optimized averaging in wireless sensor networks,” 2009 48th IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference, Shanghai, China, Dec. 2009, pp. 3763-3768. [IEEE paper→]

Abstract:
This paper studies the energy costs of running a distributed averaging/consensus algorithm in a wireless sensor network. It has recently been shown that running such an algorithm over a bidirectional spanning tree is preferable in terms of convergence time. We formulate the combinatorial optimization problem of selecting a minimal energy bidirectional spanning tree as a mixed integer linear programming problem. This problem has been shown to be NP-complete and can only be solved for small instances. We devise a semi-definite relaxation and establish bounds on the optimal cost. We also develop a series of graph-based algorithms that yield energy efficient bidirectional spanning trees and establish associated bounds on the optimal cost. Some of our algorithms can be run in a distributed manner and numerical results establish that they produce near-optimal solutions.

2008

Bin-Bin Li, Ling Wang, and Bo Liu, “An Effective PSO-Based Hybrid Algoithm for Multi-Objective Permutation Flow Shop Scheduling,” in IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, vol. 38, no. 4, pp. 818-831, July 2008. [IEEE journal→]

Abstract:
This paper proposes a hybrid algorithm based on particle swarm optimization (PSO) for a multiobjective permutation flow shop scheduling problem, which is a typical NP-hard combinatorial optimization problem with strong engineering backgrounds. Not only does the proposed multiobjective algorithm (named MOPSO) apply the parallel evolution mechanism of PSO characterized by individual improvement, population cooperation, and competition to effectively perform exploration but it also utilizes several adaptive local search methods to perform exploitation. First, to make PSO suitable for solving scheduling problems, a ranked-order value (ROV) rule based on a random key technique to convert the continuous position values of particles to job permutations is presented. Second, a multiobjective local search based on the Nawaz-Enscore-Ham heuristic is applied to good solutions with a specified probability to enhance the exploitation ability. Third, to enrich the searching behavior and to avoid premature convergence, a multiobjective local search based on simulated annealing with multiple different neighborhoods is designed, and an adaptive meta-Lamarckian learning strategy is employed to decide which neighborhood will be used. Due to the fusion of multiple different searching operations, good solutions approximating the real Pareto front can be obtained. In addition, MOPSO adopts a random weighted linear sum function to aggregate multiple objectives to a single one for solution evaluation and for guiding the evolution process in the multiobjective sense. Due to the randomness of weights, searching direction can be enriched, and solutions with good diversity can be obtained. Simulation results and comparisons based on a variety of instances demonstrate the effectiveness, efficiency, and robustness of the proposed hybrid algorithm.

Ling Wang and Bin-bin Li, “Quantum-Inspired Genetic Algorithms for Flow Shop Scheduling,” pp. 17-56. In: Nedjah N., Coelho L..S., Mourelle L..M. (eds) Quantum Inspired Intelligent Systems. Studies in Computational Intelligence, vol 121. Springer, Berlin, Heidelberg. [Springer journal→]

Introduction:
In this chapter, quantum-inspired genetic algorithms are proposed for permutation flow shop scheduling, which is a typical NP-hard combinatorial optimization problem with strong engineering background. For the singleobjective scheduling problems, a hybrid quantum-inspired genetic algorithm (HQGA) is developed to achieve more satisfactory performance. In the HQGA, a quantum-inspired GA (QGA) based on Q-bit representation is applied for exploration in discrete 0-1 hyperspace by using updating operator of quantum gate as well as genetic operators of Q-bit. And the random key representation is used to convert the Q-bit representation to job permutation for evaluating the objective value of the schedule solution. Moreover, a permutation-based GA (PGA) is applied for both performing exploration in permutation-based scheduling space and stressing exploitation for good schedule solutions. The HQGA can be viewed as a fusion of micro-space based search (Q-bit based search) and macro-space based search (permutation-based search). For the multi-objective scheduling problems, some special strategies are utilized on the basis of the HQGA for single-objective scheduling. In particular, the randomly weighted linear sum function is used in the QGA to evaluate solutions in multi-objective sense, and a non-dominated sorting technique including classification of Pareto front and fitness assignment is applied in the PGA regarding to both proximity and diversity. In addition, two trimming techniques for population are proposed and applied in the HQGA to maintain diversity of population. Simulations are carried out based on several single-objective and multi-objective benchmarks with some performance metrics. The results and comparisons demonstrate the effectiveness of the proposed hybrid quantuminspired genetic algorithms.

2007

Bin-Bin Li and Ling Wang, “A Hybrid Quantum-Inspired Genetic Algorithm for Multiobjective Flow Shop Scheduling,” in IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), vol. 37, no. 3, pp. 576-591, June 2007. [IEEE journal→]

Abstract:
This paper proposes a hybrid quantum-inspired genetic algorithm (HQGA) for the multiobjective flow shop scheduling problem (FSSP), which is a typical NP-hard combinatorial optimization problem with strong engineering backgrounds. On the one hand, a quantum-inspired GA (QGA) based on Q-bit representation is applied for exploration in the discrete 0-1 hyperspace by using the updating operator of quantum gate and genetic operators of Q-bit. Moreover, random-key representation is used to convert the Q-bit representation to job permutation for evaluating the objective values of the schedule solution. On the other hand, permutation-based GA (PGA) is applied for both performing exploration in permutation-based scheduling space and stressing exploitation for good schedule solutions. To evaluate solutions in multiobjective sense, a randomly weighted linear-sum function is used in QGA, and a nondominated sorting technique including classification of Pareto fronts and fitness assignment is applied in PGA with regard to both proximity and diversity of solutions. To maintain the diversity of the population, two trimming techniques for population are proposed. The proposed HQGA is tested based on some multiobjective FSSPs. Simulation results and comparisons based on several performance metrics demonstrate the effectiveness of the proposed HQGA.

2006

Bin-Bin Li and Ling Wang, “A Hybrid Quantum-Inspired Genetic Algorithm for Multi-objective Scheduling,” 2006 International Conference on Intelligent Computing, Kunmin, China, Aug. 2006, pp. 511-522. In: Huang DS., Li K., Irwin G.W. (eds) Intelligent Computing. Lecture Notes in Computer Science, vol. 4113 Springer, Berlin, Heidelberg.[Springer paper→]

Abstract:
This paper proposes a hybrid quantum-inspired genetic algorithm (HQGA) for multi-objective flow shop scheduling problem. On one hand, a quantum-inspired GA (QGA) based on Q-bit representation is applied for exploration in discrete 0-1 hyperspace by using updating operator of quantum gate and genetic operators of Q-bit. Random key representation is used to convert the Q-bit representation to job permutation. On the other hand, permutation-based GA (PGA) is applied for both performing exploration in permutation-based scheduling space and stressing exploitation for good schedule solutions. To evaluate solutions in multi-objective sense, randomly weighted linear sum function is used in QGA, while non-dominated sorting techniques including classification of Pareto fronts and fitness assignment are applied in PGA regarding to both proximity and diversity of solutions in multi-objective sense. Simulation results and comparisons demonstrate the effectiveness and robustness of the proposed HQGA.