Applications of Asymmetric Networks to Bio-Inspired Neural Networks for Motion Detection

2016 ◽  
pp. 231-244
Author(s):  
Naohiro Ishii ◽  
Toshinori Deguchi ◽  
Masashi Kawaguchi ◽  
Hiroshi Sasaki
Keyword(s):  
Neural Networks ◽  
Motion Detection ◽  
Asymmetric Networks

MOVEMENT AND MEMORY FUNCTION IN BIOLOGICAL NEURAL NETWORKS

1995 ◽  
Vol 04 (04) ◽  
pp. 489-500 ◽  
Author(s):  
NAOHIRO ISHII ◽  
KEN-ICHI NAKA
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Spatial Information ◽  
Memory Function ◽  
Amacrine Cells ◽  
Linear Analysis ◽  
Non Linear ◽  
Asymmetric Networks ◽  
Direction Of Movement ◽  
The Difference

Asymmetrical neural networks are shown in the biological neural network as the catfish retina. Horizontal and bipolar cell responses are linearly related to the input modulation of light, while amacrine cells work linearly and nonlinearly in their responses. These cells make asymmetrical neural networks in the retina. Several mechanisms have been proposed for the detection of motion in biological system. To make clear the difference among asymmetrical networks, we applied non-linear analysis developed by N. Wiener. Then, we can derive α-equation of movement, which shows the direction of movement. During the movement, we also can derive the movement equation, which implies that the movement holds regardless of the parameter α. By analyzing the biological asymmetric neural networks, it is shown that the asymmetric networks are excellent in the ability of spatial information processing on the retinal level. Then, the symmetric network was discussed by applying the non-linear analysis. In the symmetric neural network, it was suggested that memory function is needed to perceive the movement.

Vector Operations in Neural Network Computations

2013 ◽  
Vol 1 (2) ◽  
pp. 40-52 ◽  
Author(s):  
Naohiro Ishii ◽  
Toshinori Deguchi ◽  
Masashi Kawaguchi ◽  
Hiroshi Sasaki
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Visual Cortex ◽  
Stimulus Change ◽  
Movement Detection ◽  
Temporal Area ◽  
The Neural Network ◽  
Layered Networks ◽  
Asymmetric Networks ◽  
Odd Nonlinearity

Nonlinearity is an important factor in the biological visual neural networks. Among prominent features of the visual networks, movement detections are carried out in the visual cortex. The visual cortex for the movement detection, consist of two layered networks, called the primary visual cortex (V1), followed by the middle temporal area (MT), in which nonlinear functions will play important roles in the visual systems. These networks will be decomposed to asymmetric sub-networks with nonlinearities. In this paper, the fundamental characteristics in asymmetric and symmetric neural networks with nonlinearities are developed for the detection of the changing stimulus or the movement detection in these neural networks. By the optimization of the asymmetric networks, movement detection Equations are derived. Then, it was clarified that the even – odd nonlinearity combined asymmetric networks, has the ability of generating directional vector in the stimulus change detection or movement detection, while symmetric networks need the time memory to have the same ability. Further, the vector operations in the neural network are developed. These facts are applied to two layered networks, V1 and MT.

scholarly journals Adversarial Attacks on Multi-task Visual Perception for Autonomous Driving

2021 ◽  
Author(s):  
Ibrahim Sobh ◽  
Ahmed Hamed ◽  
Varun Ravi Kumar ◽  
Senthil Yogamani
Keyword(s):  
Neural Networks ◽  
Visual Perception ◽  
Motion Detection ◽  
Deep Neural Networks ◽  
Distance Estimation ◽  
Semantic Segmentation ◽  
Autonomous Driving ◽  
Black Box ◽  
Safety Critical ◽  
Future Work

In recent years, deep neural networks (DNNs) have accomplished impressive success in various applications, including autonomous driving perception tasks. However, current deep neural networks are easily deceived by adversarial attacks. This vulnerability raises significant concerns, particularly in safety-critical applications. As a result, research into attacking and defending DNNs has gained much coverage. In this work, detailed adversarial attacks are applied on a diverse multi-task visual perception deep network across distance estimation, semantic segmentation, motion detection, and object detection. The experiments consider both white and black box attacks for targeted and un-targeted cases, while attacking a task and inspecting the effect on all others, in addition to inspecting the effect of applying a simple defense method. We conclude this paper by comparing and discussing the experimental results, proposing insights and future work. The visualizations of the attacks are available at https://youtu.be/6AixN90budY.

Motion Detection in Asymmetric Neural Networks

2016 ◽  
pp. 409-417 ◽  
Author(s):  
Naohiro Ishii ◽  
Toshinori Deguchi ◽  
Masashi Kawaguchi ◽  
Hiroshi Sasaki
Keyword(s):  
Neural Networks ◽  
Motion Detection

Efficient Continuous-Time Asymmetric Hopfield Networks for Memory Retrieval

2010 ◽  
Vol 22 (6) ◽  
pp. 1597-1614 ◽  
Author(s):  
Pengsheng Zheng ◽  
Wansheng Tang ◽  
Jianxiong Zhang
Keyword(s):  
Neural Networks ◽  
Continuous Time ◽  
Memory Retrieval ◽  
Stable Equilibrium ◽  
Equilibrium Points ◽  
Sufficient Conditions ◽  
Asymptotically Stable ◽  
Hopfield Networks ◽  
Energy Functions ◽  
Asymmetric Networks

A novel m energy functions method is adopted to analyze the retrieval property of continuous-time asymmetric Hopfield neural networks. Sufficient conditions for the local and global asymptotic stability of the network are proposed. Moreover, an efficient systematic procedure for designing asymmetric networks is proposed, and a given set of states can be assigned as locally asymptotically stable equilibrium points. Simulation examples show that the asymmetric network can act as an efficient associative memory, and it is almost free from spurious memory problem.

Sign in / Sign up

Export Citation Format

Share Document