Realizing Behavior Level Associative Memory Learning Through Three-Dimensional Memristor-Based Neuromorphic Circuits

An, Hongyu; An, Qiyuan; Yi, Yang

doi:10.1109/tetci.2019.2921787

Cited by 35 publications

(10 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, if these two signals were presented together several times, the dogs salivated even if they only listened to the sound of a whistle with no food provided. This means the dogs can memorize the sound of whistles as a sign of food [1,6,36] through a learning/memorizing process. Through a series of experiments, Pavlov concluded that dogs have the capability of associating two originally irrelevant signals together through a training process, which is referred to as associative memory learning later.…”

Section: Associative Memory Learningmentioning

confidence: 99%

“…Numerous studies have implemented associative memory with neuromorphic systems [2,[6][7][8][9][10][11][12][13]. However, these studies merely complete a small-scale association with a few neurons in simulation environments.…”

Section: Introductionmentioning

confidence: 99%

“…Several brain regions have been proven to be involved in the learning process, including frontotemporal amygdala, hippocampus, and so on. The process of fear conditioning cannot be reproduced by other state-of-the-art associative memory models [2,[6][7][8][9][10][11][12][13] due to their limited neural network sizes. The simple neural network models cannot process informative signals, such as visual signals.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Implementation of Associative Memory Learning in Mobile Robots Using Neuromorphic Computing

Zins¹,

Zhang²,

An³

2023

Neuromorphic Computing

View full text Add to dashboard Cite

Fear conditioning is a behavioral paradigm of learning to predict aversive events. It is a form of associative learning that memorizes an undesirable stimulus (e.g., an electrical shock) and a neutral stimulus (e.g., a tone), resulting in a fear response (such as running away) to the originally neutral stimulus. The association of concurrent events is implemented by strengthening the synaptic connection between the neurons. In this paper, with an analogous methodology, we reproduce the classic fear conditioning experiment of rats using mobile robots and a neuromorphic system. In our design, the acceleration from a vibration platform substitutes the undesirable stimulus in rats. Meanwhile, the brightness of light (dark vs. light) is used for a neutral stimulus, which is analogous to the neutral sound in fear conditioning experiments in rats. The brightness of the light is processed with sparse coding in the Intel Loihi chip. The simulation and experimental results demonstrate that our neuromorphic robot successfully, for the first time, reproduces the fear conditioning experiment of rats with a mobile robot. The work exhibits a potential online learning paradigm with no labeled data required. The mobile robot directly memorizes the events by interacting with its surroundings, essentially different from data-driven methods.

show abstract

Section: Associative Memory Learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Implementation of Associative Memory Learning in Mobile Robots Using Neuromorphic Computing

Zins¹,

Zhang²,

An³

2023

Neuromorphic Computing

View full text Add to dashboard Cite

show abstract

“…With large‐scale simulations of memory synapse arrays, the proposed behavioral‐level associative memory learning system demonstrates the ability to link digital auditory and visual information together, just as our brain does. [ 31 ] In addition, as interest in ANNs increases, resistive random access memory (RRAM)‐based ANNs have also drawn attention. [ 32 ] The study of associative memory can enable ANNs to simulate more realistic memory situations, such as complex circuit designs.…”

Section: Introductionmentioning

confidence: 99%

“…[26][27][28] Application of associative memory to neural networks is multifaceted. [29][30][31][32][33] To extend the application of associative memory, two different memristor models were used in the circuit: memristor models with a forgetting effect to simulate the processing of sensory signals and memristor models with synaptic plasticity to simulate the generation and evolution of emotions. [29,30] Associative memory can also link large-scale ANNs together, not just to multiple simulated signals.…”

Section: Introductionmentioning

confidence: 99%

An Improved Memristor Model Based on the Electrochemical Metallization Effect as a Synapse for Biomimetic Applications

Dai

Wang

Yang

2021

Physica Status Solidi (b)

View full text Add to dashboard Cite

Herein, an improved memristor model with a controllable forgetting rate due to ion diffusion is proposed. A synaptic learning circuit based on this model is simulated with PSpice to complete synaptic bionics. First, the pulse pair superposition method, which is consistent with biological phenomena, is used to verify spike timing‐dependent plasticity (STDP); this proves that the modified model simulates biological synaptic behavior. Second, a habituation circuit constructed from a memristor and an analog behavioral modeling (ABM) device is designed to realize nonassociative learning. Finally, a new circuit with the improved model is built for Pavlov's associative memory implementation, simplifying the complexity of the circuit. The improved model can better realize the basic learning and memory modes of organisms and thus provide more brain‐inspired computing options.

show abstract

Memristor‐based dual mode whole‐process Pavlov associative memory circuit

Wang,

Yang,

et al. 2024

Circuit Theory & Apps

View full text Add to dashboard Cite

Pavlov associative memory experiment includes four laws: acquisition, extinction, generalization, and differentiation. The memristor synapses‐based Pavlov circuit is of great significance to the construction of neuromorphic computing system and the simulation of brain‐like function. However, few studies have been able to completely implement the whole process of Pavlov associative memory experiment. In this paper, a memristor‐based dual‐mode Pavlov whole‐process circuit is investigated and simulated; it includes the following: (1) the memristor‐based auditory neuron circuit and the visual neuron circuit are constructed to realize the learning and forgetting dual mode process of auditory and visual associative memory, respectively; (2) different from classic considering at mode switching time, the interaction between dual learning modes (the auditory learning and visual learning) is explored over a whole learning process; the whole process is proactive inhibition and retroactive inhibition in psychology; (3) different from traditional processes starting from a fixed input, a generalization and differentiation circuit is constructed to learn three similar stimulus input signals as parallel relationships and achieve the generalization and differentiation process starting from any input signal.

show abstract

Realizing Behavior Level Associative Memory Learning Through Three-Dimensional Memristor-Based Neuromorphic Circuits

Cited by 35 publications

References 39 publications

Implementation of Associative Memory Learning in Mobile Robots Using Neuromorphic Computing

Implementation of Associative Memory Learning in Mobile Robots Using Neuromorphic Computing

An Improved Memristor Model Based on the Electrochemical Metallization Effect as a Synapse for Biomimetic Applications

Memristor‐based dual mode whole‐process Pavlov associative memory circuit

Contact Info

Product

Resources

About