Baocheng Geng

Baocheng Geng

Auto-generated profile, based on publicly available information

Baocheng Geng

Department of Computer Science, The University of Alabama at Birmingham, Birmingham, AL, USA

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 April 2024 in ACM Computing Surveys
Verified
Journal Article
Open access
FZFei Zhao
Fei Zhao
CZChengcui Zhang
Chengcui Zhang
BGBaocheng Geng
Baocheng Geng
Multimodal Artificial Intelligence (Multimodal AI), in general, involves various types of data (e.g., images, texts, or data collected from different sensors), feature engineering (e.g., extraction, combination/fusion), and decision-making (e.g., majority vote). As architectures become more and more sophisticated, multimodal neural networks can integrate feature extraction, feature fusion, and decision-making processes into one single model. The boundaries between those processes are increasingly blurred. The conventional multimodal data fusion taxonomy (e.g., early/late fusion), based on which the fusion occurs in, is no longer suitable for the modern deep learning era. Therefore, based on the main-stream techniques used, we propose a new fine-grained taxonomy grouping the state-of-the-art (SOTA) models into five classes: Encoder-Decoder methods, Attention Mechanism methods, Graph Neural Network methods, Generative Neural Network methods, and other Constraint-based methods. Most existing surveys on multimodal data fusion are only focused on one specific task with a combination of two specific modalities. Unlike those, this survey covers a broader combination of modalities, including Vision + Language (e.g., videos, texts), Vision + Sensors (e.g., images, LiDAR), and so on, and their corresponding tasks (e.g., video captioning, object detection). Moreover, a comparison among these methods is provided, as well as challenges and future directions in this area.

30 January 2020 in IEEE Transactions on Signal Processing
Verified
Journal Article
Closed access
BGBaocheng Geng
Baocheng Geng
SBSwastik Brahma
Swastik Brahma
TWThakshila Wimalajeewa
Thakshila Wimalajeewa
PVPramod K. Varshney
Pramod K. Varshney
MRMuralidhar Rangaswamy
Muralidhar Rangaswamy
This paper studies human decision making via a utility based approach in a binary hypothesis testing framework that includes the consideration of individual behavioral disparity. Unlike rational decision makers who make decisions so as to maximize their expected utility, humans tend to maximize their subjective utilities, which are usually distorted due to cognitive biases. We use the value function and the probability weighting function from prospect theory to model human cognitive biases and obtain their subjective utility function in decision making. First, we show that the decision rule which maximizes the subjective utility function reduces to a likelihood ratio test (LRT). Second, to capture the unreliable nature of human decision making behavior, we model the decision threshold of a human as a Gaussian random variable, whose mean is determined by his/her cognitive bias, and the variance represents the uncertainty of the agent while making a decision. This human decision making framework under behavioral biases incorporates both cognitive biases and uncertainties. We consider several decision fusion scenarios that include humans. Extensive numerical simulations are provided throughout the paper to illustrate the impact of human behavioral biases on the performance of the decision making systems.

3 July 2020 in IEEE Transactions on Signal Processing
Verified
Journal Article
Closed access
BGBaocheng Geng
Baocheng Geng
QLQunwei Li
Qunwei Li
PVPramod K. Varshney
Pramod K. Varshney
We consider the M-ary classification problem via crowdsourcing, where crowd workers respond to simple binary questions and the answers are aggregated via decision fusion. The workers have a reject option to skip answering a question when they do not have the expertise, or when the confidence of answering that question correctly is low. We further consider that there are spammers in the crowd who respond to the questions with random guesses. Under the payment mechanism that encourages the reject option, we study the behavior of honest workers and spammers, whose objectives are to maximize their monetary rewards. To accurately characterize human behavioral aspects, we employ prospect theory to model the rationality of the crowd workers, whose perception of costs and probabilities are distorted based on some value and weight functions, respectively. Moreover, we estimate the number of spammers and employ a weighted majority voting decision rule, where we assign an optimal weight for every worker to maximize the system performance. The probability of correct classification and asymptotic system performance are derived. We also provide simulation results to demonstrate the effectiveness of our approach.

28 July 2021 in IEEE Transactions on Computational Social Systems
Verified
Journal Article
Closed access
BGBaocheng Geng
Baocheng Geng
XCXiancheng Cheng
Xiancheng Cheng
SBSwastik Brahma
Swastik Brahma
DKDavid Kellen
David Kellen
PVPramod K. Varshney
Pramod K. Varshney
While there has been extensive work on modeling of human decision-making both for individuals and groups from a cognitive psychology point of view, research on this topic from a signal processing and information fusion perspective is relatively recent. In this work, we consider a distributed detection problem consisting of a number of human local decision makers and a fusion center (FC). Signal detection theory is exploited to answer why promoting heterogeneity could improve the performance of collaborative human decision-making. We consider the following two scenarios: 1) the local decision makers are independent and the level of heterogeneity is measured in terms of the variability of human expertise and 2) humans make correlated local decisions due to their perceptual and behavioral similarities and heterogeneity is measured by the amount of correlation. In both cases, we show that the detection performance of the FC can be improved with the increase of heterogeneity. In particular, in the second scenario, we develop a portfolio theory-based framework to select participants from correlated human agents so that heterogeneity is enhanced resulting in improved decision-making performance. Simulations are provided for illustration and performance comparison.

7 January 2022 in IEEE Transactions on Signal and Information Processing over Networks
Verified
Journal Article
Closed access
CQChen Quan
Chen Quan
BGBaocheng Geng
Baocheng Geng
YHYunghsiang S. Han
Yunghsiang S. Han
PVPramod K. Varshney
Pramod K. Varshney
This paper employs an audit bit based mechanism to mitigate the effect of Byzantine attacks on distributed Bayesian detection systems. In this framework, the optimal attacking strategy for strategic attackers is investigated for the traditional audit bit based scheme (TAS) to evaluate the robustness of the system. We show that it is possible for a strategic attacker to degrade the performance of TAS to the system without audit bits. To enhance the robustness of the system in the presence of strategic attackers, we propose an enhanced audit bit based scheme (EAS). The optimal fusion rule for the proposed scheme is derived and the detection performance of the system is evaluated via the probability of error for the system. Simulation results show that the proposed EAS improves the robustness and the detection performance of the system. Moreover, based on EAS, another new scheme called the reduced audit bit based scheme (RAS) is proposed which further improves system performance. We derive the new optimal fusion rule and the simulation results show that RAS outperforms EAS and TAS in terms of both robustness and detection performance of the system. Then, we extend the proposed RAS for a wide-area cluster based distributed wireless sensor networks (CWSNs). Simulation results show that the proposed RAS significantly reduces the communication overhead between the sensors and the FC, which prolongs the lifetime of the network.

1 December 2023 in IEEE Transactions on Signal Processing
Verified
Journal Article
Closed access
CQChen Quan
Chen Quan
YHYunghsiang S. Han
Yunghsiang S. Han
BGBaocheng Geng
Baocheng Geng
PVPramod K. Varshney
Pramod K. Varshney
This paper investigates distributed detection of sparse stochastic signals with quantized measurements under Byzantine attacks, where sensors may send falsified data to the Fusion Center (FC) to degrade system performance. Here, the Bernoulli-Gaussian (BG) distribution is used to model sparse stochastic signals. Several detectors with significantly improved detection performance are proposed by incorporating estimates of attack parameters into the detection process. In the case of unknown sparsity degree and attack parameters, we propose the generalized likelihood ratio test with reference sensors (GLRTRS) as well as the locally most powerful test with reference sensors (LMPTRS). Our simulation results show that these detectors outperform the LMPT and GLRT detectors designed in attack-free environments and achieve detection performance close to the benchmark likelihood ratio test (LRT) detector. In the case of unknown sparsity degree and known fraction of Byzantine nodes in the network, we further propose enhanced LMPTRS (E-LMPTRS) and enhanced GLRTRS (E-GLRTRS) detectors by filtering out potential malicious sensors in the network, resulting in improved detection performance compared to GLRTRS and LMPTRS detectors.