11-09-2014, 11:12 AM
ARTIFICLAL NEURAL NETWORK BIDIRECTIONAL ASSOCIATIVE MEMORY PROJECT REPORT
ARTIFICLAL NEURAL NETWORK.doc (Size: 50.5 KB / Downloads: 10)
ABSTRACT
This paper focuses on the bidirectional associative memory its features and the future aspects and the current context of application BAM is a type of neural network. Artificial neural network (Ann’s) resembled the human nervous system, with algorithms consisting of weighted interconnecting processing units (like neural map of the human brain). To address a particular problem using Ann’s, the interrelated connections are tuned and the value of weights between units is needed. Neural network is a new unexplored topic of interest for the computer scientists.
Bam comes under recurrent types of network called Hopfield network? BAM is a resonance model, in the sense that information is passed back and forth between two layers of units until a stable sate is reached. The Hopfield network is said to be auto associative, because it uses a partial and noisy pattern to recall the best match of itself.
Applications of BAM include: 1ASSOCIATIVE NEURAL MEMORIES:
Associative neural memories are a class of artificial neural networks (connectionist nets) which have gained substantial attention relative to other neural net paradigms. Associative memories have been the subject of research 2) NOISE TOLERANCY:
This paper analyzes the sensitivity to noise in BAM (Bidirectional Associative Memory), and then proves the noise immunity of BAM relates not only to the minimum absolute value of net inputs (MAV) but also to the variance of weights associated with synapse connections
Case studies about the NEURAL COMPUTATION: - Many scientists and engineers now use neural networks to tackle problems that are either intractable or unrealistically time consuming to solve, through traditional computational strategies. FAULT TOLERANT SYSTEMS. Recent advances in computer technology have made the design of large and very flexible associative processors possible.
The future aspects of BAM such as DATABASE MANAGEMENT SYSTEM and NATURAL LANGUAGE LEARNING THROUGH ABSTRACT MEMORY and LIFE AND ROBOTICs’’’’ are also featured.
NEURAL NETWORKS BIDIRECTIONAL ASSOCIATIVE MEMORY
Artificial neural networks are intended for modeling the organizational principles of
Central nervous system. So that ANN will allow cognitive and sensory tasks to be performed much faster than is possible using conventional serial processors.
ANN is an information processing paradigm that is inspired by the way biological nervous systems, such as the brain, process information. The key element of this paradigm is the novel structure of the information processing system. It is composed of a large number of highly interconnected processing elements (neurons) working in unison to solve specific problems..
There are six types of neural networks one among them is BAM-Bidirectional associative memory. Which comes under recurrent type of network called as Hopfield network? The bidirectional associative memory can be viewed as a generalization of the Hopfield model, to allow for a heteroassociative memory to be implemented. In this case, the association is between names and corresponding phone numbers.
The bidirectional associative memory is related to the e Hopfield network and also has some similarity to the ART architecture. It consists of two layers. Uses the forward and backward information flow between the layers to perform a search for a stored stimulus
Features of BAM
Kosko (1988) extended the Hopfield model by incorporating an additional layer to perform recurrent auto associations as well as heteroassociations on the stored memories.
The network structure of the Bidirectional Associative Memory model is similar to that of the linear associatoror but the connections are bidirectional, i.e., wij = wji, for i = 1, 2, ..., m and j = 1, 2, ..., n. Also, the units in both layers serve as both input and output units depending on the direction of propagation. Propagating signals from the X layer to the Y layer makes the units in the X layer act as input units while the units in the Y layer act as output units. The same is true for the other direction, i.e., propagating from the Y layer to the X layer makes the units in the Y layer act as input units while the units in the X layer act as output units
APPLICATIONS OF BAM
Associative Neural Memories
Associative neural memories are a class of artificial neural networks (connectionist nets) which have gained substantial attention relative to other neural net paradigms. Associative memories have been the subject of research since the early seventies. Recent interest in these memories has been spurred by the seminal work of Hopfield in the early eighties, who has shown how a simple discrete nonlinear dynamical system can exhibit associative recall of stored binary patterns through collective computing. Since then, a number of important contributions have appeared in conference proceedings and technical journals addressing various issues of associative neural memories, including multiple-layer architectures, recording/storage algorithms, capacity, retrieval dynamics, fault-tolerance, and hardware implementation.
Noise tolerance:
This application analyzes the sensitivity to noise in BAM (Bidirectional Associative Memory), and then proves the noise immunity of BAM relates not only to the minimum absolute value of net inputs (MAV) but also to the variance of weights associated with synapse connections. In fact, it is a positive monotonically increasing function of the quotient of MAV divided by the variance of weights. Besides, the performance of pseudo-relaxation method depends on learning parameters (lambda and xi), but the relation of them is not linear. So it is hard to find a best combination of lambda and xi which leads to the best BAM performance. And it is obvious that pseudo-relaxation is a kind of local optimization method, so it cannot guarantee to get the global optimal solution.
FUTURE ASPECTS OF BAM
Database Management:
Its broad coverage of expert systems, decision support, artificial neural networks, fuzzy systems and evolutionary computation will show how intelligent systems work together and be utilized in health care and the public health practice. Some applications include networks that perform: -Pattern
Life and Robotics
This covers a broad muArtificial Life and Robotics is an international journal publishing original technical papers and authoritative state-of-the-art reviews on the development of new technologies concerning artificial life and robotics, especially computer-based simulation and hardware for the twenty-first centuryltidisciplinary field, including areas such as artificial brain research, artificial intelligence, artificial life, artificial living, artificial mind research, chaos, cognitive science, complexity, evolutionary computations, fuzzy control, genetic algorithms, innovative computations, micro machines, mobile vehicles, neural networks, neurocomputers, neurocomputing technologies and applications, robotics, and virtual reality. Hardware-oriented submissions are particularly welcome.
CONCLUSION
Neural networks is an .Expectations of faster and better solutions provide us with the challenges to build machines using the same computations and organizational principles, simplified and abstracted from neurobiological studies of the brain. BAM is a resonance model, in the sense that information is passed back and forth between two layers of units until a stable sate is reached.Thus BAM can be used to analyse and understand the complex problem domains the connection of current problem can be solved using previous experiences of the memory of the system. Thus can be of great interest to computer scientist to develop software to imitate the human brains capacity to recollect information from past experiences.