Error tolerant associative memory

Liou, Cheng-Yuan; Yuan, Shao-Kuo

doi:10.1007/s004220050566

Cited by 14 publications

(12 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, the pseudo-inverse rule (also called projection rule) can increase HNNs storage capacity and improve accuracy (Wu et al, 2012;Sahoo et al, 2016), but is not local nor incremental. Moreover, recent works have explored learning rules with self-feedback connections (non-0 diagonal), and have shown higher accuracy for a high number of stored patterns (Liou and Yuan, 1999;Folli et al, 2017;Rocchi et al, 2017;Gosti et al, 2019). In summary, despite the present limitations of the ONN, features in terms of FPS, computation time and training, are encouraging toward the exploration of a wider range of applications.…”

Section: Limitations and Future Directionsmentioning

confidence: 99%

Digital Implementation of Oscillatory Neural Network for Image Recognition Applications

et al. 2021

View full text Add to dashboard Cite

Computing paradigm based on von Neuman architectures cannot keep up with the ever-increasing data growth (also called “data deluge gap”). This has resulted in investigating novel computing paradigms and design approaches at all levels from materials to system-level implementations and applications. An alternative computing approach based on artificial neural networks uses oscillators to compute or Oscillatory Neural Networks (ONNs). ONNs can perform computations efficiently and can be used to build a more extensive neuromorphic system. Here, we address a fundamental problem: can we efficiently perform artificial intelligence applications with ONNs? We present a digital ONN implementation to show a proof-of-concept of the ONN approach of “computing-in-phase” for pattern recognition applications. To the best of our knowledge, this is the first attempt to implement an FPGA-based fully-digital ONN. We report ONN accuracy, training, inference, memory capacity, operating frequency, hardware resources based on simulations and implementations of 5 × 3 and 10 × 6 ONNs. We present the digital ONN implementation on FPGA for pattern recognition applications such as performing digits recognition from a camera stream. We discuss practical challenges and future directions in implementing digital ONN.

show abstract

Section: Limitations and Future Directionsmentioning

confidence: 99%

Digital Implementation of Oscillatory Neural Network for Image Recognition Applications

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Several kinds of AM have been implemented. Among these are Bidirectional AM (BAM) [9], Morphological AM (MAM) [10], multi associative neural network (MANN) [11], error Tolerant Associative Memory (eTAM) [12], hetero-AM [13], Theory of Distributed Associative Memory (TODAM) [14] and Recurrent Correlation Associative Memory (RCAM) [15].…”

Section: Introductionmentioning

confidence: 99%

“…TODAM uses a vector type memory which has the risk of suffering from the crosstalk problem [16]. Determining weights to produce correct output results in BAM [9], MANN [11], and eTAM [12] is time consuming . The kernel image that has to be used in the recall process in MAM [10] and RCAM [15] is difficult to built.…”

Section: Introductionmentioning

confidence: 99%

Multi-wheel graph neuron

Amir

Khan

Mahmood

2009

Proceedings of the 11th International Conference on Information Integration and Web-Based Applications &Amp; Services

View full text Add to dashboard Cite

The significant growing amount of shared files in distributed and serverless P2P networks has impacted heavily on content management and retrieval field. In this paper, we propose an associative memory network with intention to provide efficient pattern recognition within P2P networks. An efficient pattern recognition algorithm which works within distributed and dynamic nature of P2P network would be beneficial for better content handling and distribution. Our approach, which is called multi-wheel Graph Neuron (mWGN), is a variant of Graph Neuron (GN)-based algorithm built on top of Chord overlay network. GN algorithm is an associative memory designed for wireless sensor environment. mWGN is a restructured design from Hierarchical Graph Neuron (HGN), preserving its single cycle learning, lightweight and accuracy feature while requiring fewer number of nodes. Result from the experiment shows the proposed approach is highly accurate and the fault tolerance of Chord protocol provides stability for the proposed approach.

show abstract

“…There is a perfect design for the hidden neurons in (Liou and Sou, 2003). The design in (Liou and Yuan, 1999) provides a biologically plausible solution for the tolerance ability without any hidden neuron and annealing process. This work will present a method to further explore the idea behind this design and formulate it for finite memory loading.…”

Section: Introductionmentioning

confidence: 99%

“…The goal is somewhat similar to that of Gardner for extensive memory loading in (Gardner, 1989). This work will use the same experimental simulations as those in (Liou and Yuan, 1999) to ease comparison.…”

Section: Introductionmentioning

confidence: 99%

Finite Memory Loading in Hairy Neurons

Liou

Lin

2006

Nat Comput

Self Cite

View full text Add to dashboard Cite

This paper presents a method to expand the basins of stable patterns in associative memory. It examines fully-connected associative memory geometrically and translate the learning process into an algebraic optimization procedure. It finds that locating all the patterns at certain stable corners of the neurons' hypercube as far from the decision hyperplanes as possible can produce excellent error tolerance. It then devises a method based on this finding to develop the hyperplanes. This paper further shows that this method leads to the hairy model, or the deterministic analogue of the Gibb's free energy model. Through simulations, it shows that this method gives better error tolerance than does the Hopfield model and the error-correction rule in both synchronous and asynchronous modes.

show abstract

Error tolerant associative memory

Cited by 14 publications

References 27 publications

Digital Implementation of Oscillatory Neural Network for Image Recognition Applications

Digital Implementation of Oscillatory Neural Network for Image Recognition Applications

Multi-wheel graph neuron

Finite Memory Loading in Hairy Neurons

Contact Info

Product

Resources

About