IEEE Computational Intelligence Magazine - November 2020 - 7

agents; ranging from A* search to stateof-the-art deep reinforcement learning
(RL). Furthermore, we show that the
learning potential of state-of-the-art
deep RL models does not seamlessly
transfer from the benchmark environ-
ments to target ones without heavily
tuning their hyperparameters, leading to
linear scaling of the engineering efforts,
and computational cost with the num-
ber of target domains."
IEEE Transactions on Cognitive
and Developmental Systems

Concrete Action Representation Model:
From Neuroscience to Robotics, by J.
Nassour, T. D. Hoa, P. Atoofi, and F.
Hamker, IEEE Transactions on Cognitive and Developmental Systems,Vol. 12,
No. 2, June 2020, pp. 272-284.
Digital Object Identifier: 10.1109/
TCDS.2019.2896300
"How can robotics benefit from
neuroscience to build a unified frame-
work that computes actions for both
locomotion and manipulation tasks?
Inspired by the hierarchical neural con-
trol of movement from cortex to spinal
cord, the authors propose a model that
generates a concrete action representa-
tion in robotics. The action program is
composed of four basic modules: 1) pat-

tern selection; 2) spatial coordination; 3)
temporal coordination; and 4) sensory
motor adaptation. The first and the
fourth are considered for behavior initi-
ation. The model is implemented on a
humanoid robot to generate rhythmic
and nonrhythmic movements. The robot
is able to perform tasks like perturbation
recovery, and drawing based on different
motor programs generated by the same
model. Unifying motor control in
robotics through a hierarchical structure
increases the capacity to gain an accurate
and deep understanding of transfer of
motor skills between different tasks."
IEEE Transactions on Emerging
Topics in Computational
Intelligence

Pedestrian Flow Optimization to Reduce
the Risk of Crowd Disasters Through
Human-Robot Interaction, by C. Jiang,
Z. Ni, Y. Guo, and H. He, IEEE
Transactions on Emerging Topics in
Computational Intelligence, Vol. 4, No.
3, June 2020, pp. 298-311.
Digital Object Identifier: 10.1109/
TETCI.2019.2930249
"Pedestrian flow in densely populated
or congested areas usually presents irreg-
ular or turbulent motion state due to
competitive behaviors of individual

pedestrians, which reduces flow efficien-
cy and raises the risk of crowd accidents.
Effective pedestrian flow regulation strat-
egies are highly valuable for flow optimi-
zation. Existing studies seek for optimal
design of indoor architectural features
and spatial placement of pedestrian facil-
ities for the purpose of flow optimiza-
tion. However, once placed, the stationary
facilities are not adaptive to real-time
flow changes. In this paper, we investi-
gate the problem of regulating two
merging pedestrian flows in a bottleneck
area using a mobile robot moving
among the pedestrian flows. The pedes-
trian flows are regulated through dynam-
ic human-robot interaction (HRI)
during their collective motion. We adopt
an adaptive dynamic programming (ADP)
method to learn the optimal motion
parameters of the robot in real time, and
the resulting outflow through the bottle-
neck is maximized with the crowd pres-
sure reduced to avoid potential crowd
disasters. The proposed algorithm is a
data-driven approach that only uses
camera observation of pedestrian flows
without explicit models of pedestrian
dynamics and HRI. Extensive simula-
tion studies are performed in both
MATLAB and a robotic simulator to
verify the proposed approach and evalu-
ate the performances."

Call for Papers for Journal Special Issues
Special Issue on "Effective Feature Fusion in Deep Neural Networks"

Journal: IEEE Transactions on Neural Networks and Learning Systems
Guest Editors: Yanwei Pang (pyw@tju.edu.cn), Fahad Shahbaz Khan, Xin Lu, and Fabio Cuzzolin
Submission Deadline: November 30, 2020
https://cis.ieee.org/images/files/Documents/call-for-papers/tnnls/SI_EFDNN_TNNLS_CFP.pdf

Special Issue on "Deep Learning for Anomaly Detection"

Journal: IEEE Transactions on Neural Networks and Learning Systems
Guest Editors: Guansong Pang (guansong.pang@adelaide.edu.au), Charu Aggarwal, Chunhua Shen, and
Nicu
Submission Deadline: November 30, 2020
https://cis.ieee.org/images/files/Documents/call-for-papers/tnnls/TNNLS_SI_deep_learning_for_anomaly_detecti
on_CFP.pdf

Digital Object Identifier 10.1109/MCI.2020.3019905

NOVEMBER 2020 | IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE

https://cis.ieee.org/images/files/Documents/call-for-papers/tnnls/SI_EFDNN_TNNLS_CFP.pdf https://cis.ieee.org/images/files/Documents/call-for-papers/tnnls/TNNLS_SI_deep_learning_for_anomaly_detection_cpf.pdf https://cis.ieee.org/images/files/Documents/call-for-papers/tnnls/TNNLS_SI_deep_learning_for_anomaly_detection_cpf.pdf

IEEE Computational Intelligence Magazine - November 2020

Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - November 2020