A5062908679- Advanced Memory and Neural Computing
- Neural dynamics and brain function
- CCD and CMOS Imaging Sensors
- Neuroscience and Neural Engineering
- Anomaly Detection Techniques and Applications
- Generative Adversarial Networks and Image Synthesis
- Evolutionary Algorithms and Applications
- Adversarial Robustness in Machine Learning
- Neural Networks and Applications
Huazhong University of Science and Technology
2021-2025
Ministry of Education of the People's Republic of China
2023
In this work, an operant conditioning (OC) model and memristive circuit implementation based on emotion generation modulation is proposed, which inspired by neural psychological mechanisms to simulate human behavioral decisions guided reinforcers emotional changes during interaction with dynamically changing environments. The OC consists of four main modules are the cue-synapse-action module, reward punishment value systems, system. Under conditions appearance or termination, designed...
Pavlov conditioning is a typical associative memory, which involves learning between the gustatory and auditory cortex, known as memory. Inspired by neural mechanisms biological phenomena of this paper proposes multi-functional memristive memory circuit. In addition to forgetting, whose rates change with number times, circuit also achieves other innovative functions. First, consolidation learning, refers continued process after acquiring changes forgetting. Secondly, natural forgetting rate...
In this work, we propose a multisensory mutual associative memory networks framework and memristive circuit to mimic the ability of biological brain make associations information received simultaneously. The inspired by neural mechanisms cells mainly consists three modules: 1) storage neurons module, which encodes external multimodal into firing rate spikes; 2) synapse uses nonvolatility memristor achieve weight adjustment learning; 3) retrieval neuron feeds signal output from each sensory...
Although deep learning of 3D point clouds has made significant progress, the robustness models not been fully investigated. Existing attack methods show satisfactory performance under white-box setting but frequently suffer from a low transferability to black-box models. A key improve attacks is find commonalities multiple To this end, novel method proposed by only leveraging perturbations in local geometric feature areas termed FAP. Specifically, are pertinently added calculating curvature...
Affective associative memory is one method by which agents acquire knowledge, experience, and skills from natural surroundings or social activities. Using neuromorphic circuits to implement affective aids in developing brain-inspired intelligence. In this paper, a feature-affective (FAAM) network model its memristive circuit are proposed for real-time mutual retrieval between multiple features emotions. With the context of fear conditioning, FAAM verified enable acquisition extinction...