Abstract
A mixed order hyper network (MOHN) is a neural network in which weights can connect any number of neurons, rather than the usual two. MOHNs can be used as content addressable memories (CAMs) with higher capacity than standard Hopfield networks. They can also be used for regression learning of functions in \(f:\{-1,1\}^n \rightarrow \mathbb {R}\) in which the turning points are equivalent to memories in a CAM. This paper presents a number of methods for learning an energy function from data that can act as either a CAM or a regression model and presents the advantages of using such an approach.
This is a preview of subscription content, log in via an institution.
Notes
- 1.
The regression Eq. 9 is actually the negative of the energy function, which is minimised by applying the settling Algorithm 2.
References
Beauchamp, K.: Applications of Walsh and Related Functions. Academic Press, London (1984)
Caparrós, G.J., Ruiz, M.A.A., Hernández, F.S.: Hopfield neural networks for optimization: study of the different dynamics. Neurocomputing 43(1–4), 219–237 (2002)
Hastie, T., Tibshirani, R., Friedman, J.: The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Springer, New York (2009)
Heckendorn, R.B., Wright, A.H.: Efficient linkage discovery by limited probing. Evol. Comput. 12(4), 517–545 (2004)
Hopfield, J.J., Tank, D.W.: Neural computation of decisions in optimization problems. Biol. Cybern. 52, 141–152 (1985)
Hopfield, J.J.: Neural networks and physical systems with emergent collective computational abilities. Proc. Nat. Acad. Sci. USA 79(8), 2554–2558 (1982)
Kubota, T.: A higher order associative memory with Mcculloch-Pitts neurons and plastic synapses. In: 2007 International Joint Conference on Neural Networks, IJCNN 2007, pp. 1982–1989, August 2007
Pelikan, M., Goldberg, D.E., Cantú-paz, E.E.: Linkage problem, distribution estimation, and Bayesian networks. Evol. Comput. 8(3), 311–340 (2000)
Storkey, A.J., Valabregue, R.: The basins of attraction of a new hopfield learning rule. Neural Netw. 12(6), 869–876 (1999)
Swingler, K.: Applying Neural Networks: A Practical Guide. Academic Press, London (1996)
Swingler, K.: On the capacity of Hopfield neural networks as EDAs for solving combinatorial optimisation problems. In: Proceedings of IJCCI (ECTA), pp. 152–157. SciTePress (2012)
Swingler, K.: A comparison of learning rules for mixed order hyper networks. In: Proceedings of IJCCI (NCTA) (2015)
Swingler, K.: Local optima suppression search in mixed order hyper networks. In: 2015 15th UK Workshop on Computational Intelligence (UKCI) (2015)
Swingler, K.: Opening the black box: analysing MLP functionality using walsh functions. In: Merelo, J.J., Rosa, A., Cadenas, J.M., Dourado, A., Madani, K., Filipe, J. (eds.) Computational Intelligence. Studies in Computational Intelligence, vol. 620, pp. 303–323. Springer, Cham (2016)
Swingler, K.: Structure discovery in mixed order hyper networks. Big Data Analytics 1(1), 8 (2016)
Swingler, K., Smith, L.: Training and making calculations with mixed order hyper-networks. Neurocomputing 141, 65–75 (2014)
Swingler, K., Smith, L.: An analysis of the local optima storage capacity of hopfield network based fitness function models. In: Nguyen, N.T., Kowalczyk, R., Fred, A., Joaquim, F. (eds.) Transactions on Computational Collective Intelligence XVII. LNCS, vol. 8790, pp. 248–271. Springer, Heidelberg (2014). doi:10.1007/978-3-662-44994-3_13
Tibshirani, R.: Regression shrinkage and selection via the lasso. J. Roy. Stat. Soc. Series B (Methodological), 267–288 (1996)
Venkatesh, S.S., Baldi, P.: Programmed interactions in higher-order neural networks: maximal capacity. J. Complex. 7(3), 316–337 (1991)
Walsh, J.: A closed set of normal orthogonal functions. Amer. J. Math. 45, 5–24 (1923)
Wilson, G.V., Pawley, G.S.: On the stability of the travelling salesman problem algorithm of hopfield and tank. Biol. Cybern. 58(1), 63–70 (1988)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Swingler, K. (2017). High Capacity Content Addressable Memory with Mixed Order Hyper Networks. In: Merelo, J.J., et al. Computational Intelligence. IJCCI 2015. Studies in Computational Intelligence, vol 669. Springer, Cham. https://doi.org/10.1007/978-3-319-48506-5_17
Download citation
DOI: https://doi.org/10.1007/978-3-319-48506-5_17
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-48504-1
Online ISBN: 978-3-319-48506-5
eBook Packages: EngineeringEngineering (R0)