A DISCRETE FULLY RECURRENT NETWORK OF MAX PRODUCT UNITS FOR ASSOCIATIVE MEMORY AND CLASSIFICATION

2002 ◽  
Vol 12 (03n04) ◽  
pp. 247-262 ◽  
Author(s):  
ROELOF K. BROUWER
Keyword(s):  
Fixed Points ◽  
Associative Memory ◽  
Recurrent Network ◽  
Connection Matrix ◽  
Recurrent Networks ◽  
Associative Memories ◽  
Product Operation ◽  
Linear Threshold ◽  
Fully Connected

This paper defines the truncated normalized max product operation for the transformation ofstates of a network and provides a method for solving a set of equations based on this operation. The operation serves as the transformation for the set of fully connected units in a recurrent network that otherwise might consist of linear threshold units. Component values of the state vector and ouputs of the units take on the values in the set {0, 0.1, …, 0.9, 1}. The result is a much larger state space given a particular number of units and size of connection matrix than for a network based on threshold units. Since the operation defined here can form the basis of transformations in a recurrent network with a finite number of states, fixed points or cycles are possible and the network based on this operation for transformations can be used as an associative memory or pattern classifier with fixed points taking on the role of prototypes. Discrete fully recurrent networks have proven themselves to be very useful as associative memories and as classifiers. However they are often based on units that have binary states. The effect of this is that the data to be processed consisting of vectors in ℜn have to be converted to vectors in {0, 1}m with m much larger than n since binary encoding based on positional notation is not feasible. This implies a large increase in the number of components. The effect can be lessened by allowing more states for each unit in our network. The network proposed demonstrates those properties that are desirable in an associative memory very well as the simulations show.

A DISCRETE FULLY RECURRENT NETWORK OF SIGMOID UNITS FOR ASSOCIATIVE MEMORY AND PATTERN CLASSIFICATION

2002 ◽  
Vol 16 (05) ◽  
pp. 527-550
Author(s):  
ROELOF K. BROUWER
Keyword(s):  
Associative Memory ◽  
Pattern Classification ◽  
Limit Cycles ◽  
Learning Algorithms ◽  
Weight Vector ◽  
Recurrent Network ◽  
Connection Matrix ◽  
Recurrent Networks ◽  
Associative Memories ◽  
Effective Learning

Fully recurrent networks have proven themselves to be very useful as associative memories and as classifiers. However, they are generally based on units that have binary states. The effect of this is that data to be processed consisting of vectors in [Formula: see text] have to be converted to vectors in {0, 1}mwith m much larger than n since binary encoding based on positional notation is not feasible. This implies a large increase in number of components. This effect can be lessened by allowing more states for each unit in our network.This paper describes two effective learning algorithms for a network whose units take the dot product of the input with a weight vector, followed by a tanh transformation and a discretization transformation in the form of rounding or truncation. The units have states that are in {0, 0.1, 0.2, …, 0.9, 1} rather than in {0, 1} or {-1, 1}. The result is a much larger state space given a particular number of units and size of connection matrix. Two convergent learning algorithms for training such a network to store fixed points or attractors are proposed. The network exhibits those properties that are desirable in an associative memory such as limit cycles of 1, attraction to the closest attractor and few transitions required to reach attractors. Since memories that are stored can be used to represent prototypes of patterns the network is useful for pattern classification. A pattern to be classified would be entered and its class would be the same as the class of the prototype to which it is attracted to which it is.

scholarly journals The Role of Visual Association Cortex in Associative Memory Formation across Development

2018 ◽  
Vol 30 (3) ◽  
pp. 365-380 ◽  
Author(s):  
Maya L. Rosen ◽  
Margaret A. Sheridan ◽  
Kelly A. Sambrook ◽  
Matthew R. Peverill ◽  
Andrew N. Meltzoff ◽  
...  
Keyword(s):  
Associative Memory ◽  
Visual Processing ◽  
Memory Performance ◽  
Neural Mechanism ◽  
Association Cortex ◽  
Associative Memories ◽  
Associate Learning ◽  
Visual Association Cortex ◽  
Age Related

Associative learning underlies the formation of new episodic memories. Associative memory improves across development, and this age-related improvement is supported by the development of the hippocampus and pFC. Recent work, however, additionally suggests a role for visual association cortex in the formation of associative memories. This study investigated the role of category-preferential visual processing regions in associative memory across development using a paired associate learning task in a sample of 56 youths (age 6–19 years). Participants were asked to bind an emotional face with an object while undergoing fMRI scanning. Outside the scanner, participants completed a memory test. We first investigated age-related changes in neural recruitment and found linear age-related increases in activation in lateral occipital cortex and fusiform gyrus, which are involved in visual processing of objects and faces, respectively. Furthermore, greater activation in these visual processing regions was associated with better subsequent memory for pairs over and above the effect of age and of hippocampal and pFC activation on performance. Recruitment of these visual processing regions mediated the association between age and memory performance, over and above the effects of hippocampal activation. Taken together, these findings extend the existing literature to suggest that greater recruitment of category-preferential visual processing regions during encoding of associative memories is a neural mechanism explaining improved memory across development.

scholarly journals A Study of Associative Memories with Hopfield neural network Model for handwritten character recognition

2018 ◽  
Vol 3 (01) ◽  
Author(s):  
Sandeep Kumar ◽  
Manu Pratap Singh
Keyword(s):  
Neural Network ◽  
Associative Memory ◽  
Network Model ◽  
Neural Network Model ◽  
Character Recognition ◽  
Hopfield Neural Network ◽  
Recurrent Network ◽  
Associative Memories ◽  
Bidirectional Associative Memory ◽  
Content Addressable Memory

Neural network is the most important model which has been studied in past decades by several researchers. Hopfield model is one of the network model proposed by J.J. Hopfield that describes the organization of neurons in such a way that they function as associative memory or also called content addressable memory. This is a recurrent network similar to recurrent layer of the hamming network but which can effectively perform the operation of both layer hamming network. The design of recurrent network has always been interesting problems to research and a lot of work is going on present application. In present paper we will discuss about the design of Hopfield Neural Network (HNNs), bidirectional associative memory (BAMs) and multidirectional associative memory (MAMs) for handwritten characters recognition. Recognized characters are Hindi alphabets.

A Solution Method for Truncated Normalized Max Product Fuzzy Set of Equations

2003 ◽  
Vol 11 (03) ◽  
pp. 361-372
Author(s):  
Roelof K. Brouwer
Keyword(s):  
Iterative Method ◽  
Finite Number ◽  
Fuzzy Set ◽  
State Vector ◽  
Hardware Implementation ◽  
Solution Method ◽  
Recurrent Network ◽  
The State ◽  
Linear Threshold ◽  
Fully Connected

An iterative method of solving a set of equations based on the truncated normalized max product is described. The operation may serve as the transformation for the set of fully connected units in a fully recurrent network that might otherwise consist of linear threshold units. Because of truncation and normalization the network acting under this transformation has a finite number of states and components of the state vector are bounded. Component values however are not restricted to binary values as would be the case if the network consisted of linear threshold units but can now take on the values in the set {0, 0.1,..0.9, 1}. This means that each unit although still having discrete output can provide finer granularity compared to the case where a linear threshold unit is used. Truncation is natural in hardware implementation where only a finite number of places behind the decimal are retained.

scholarly journals GateRL: Automated Circuit Design Framework of CMOS Logic Gates Using Reinforcement Learning

Electronics ◽  
2021 ◽  
Vol 10 (9) ◽  
pp. 1032
Author(s):  
Hyoungsik Nam ◽  
Young In Kim ◽  
Jina Bae ◽  
Junhee Lee
Keyword(s):  
Reinforcement Learning ◽  
Circuit Design ◽  
Logic Gates ◽  
Connection Matrix ◽  
Design Framework ◽  
Network Layers ◽  
Learning Speed ◽  
Fully Connected ◽  
Automated Circuit Design ◽  
Circuit Configuration

This paper proposes a GateRL that is an automated circuit design framework of CMOS logic gates based on reinforcement learning. Because there are constraints in the connection of circuit elements, the action masking scheme is employed. It also reduces the size of the action space leading to the improvement on the learning speed. The GateRL consists of an agent for the action and an environment for state, mask, and reward. State and reward are generated from a connection matrix that describes the current circuit configuration, and the mask is obtained from a masking matrix based on constraints and current connection matrix. The action is given rise to by the deep Q-network of 4 fully connected network layers in the agent. In particular, separate replay buffers are devised for success transitions and failure transitions to expedite the training process. The proposed network is trained with 2 inputs, 1 output, 2 NMOS transistors, and 2 PMOS transistors to design all the target logic gates, such as buffer, inverter, AND, OR, NAND, and NOR. Consequently, the GateRL outputs one-transistor buffer, two-transistor inverter, two-transistor AND, two-transistor OR, three-transistor NAND, and three-transistor NOR. The operations of these resultant logics are verified by the SPICE simulation.

Recurrent Neural Networks with Small Weights Implement Definite Memory Machines

2003 ◽  
Vol 15 (8) ◽  
pp. 1897-1929 ◽  
Author(s):  
Barbara Hammer ◽  
Peter Tiňo
Keyword(s):  
Neural Networks ◽  
Recurrent Neural Networks ◽  
Experimental Studies ◽  
Activation Function ◽  
Transition Function ◽  
Recurrent Network ◽  
Point Of View ◽  
Recurrent Networks ◽  
Contraction Parameter ◽  
Arbitrary Precision

Recent experimental studies indicate that recurrent neural networks initialized with “small” weights are inherently biased toward definite memory machines (Tiňno, Čerňanský, & Beňušková, 2002a, 2002b). This article establishes a theoretical counterpart: transition function of recurrent network with small weights and squashing activation function is a contraction. We prove that recurrent networks with contractive transition function can be approximated arbitrarily well on input sequences of unbounded length by a definite memory machine. Conversely, every definite memory machine can be simulated by a recurrent network with contractive transition function. Hence, initialization with small weights induces an architectural bias into learning with recurrent neural networks. This bias might have benefits from the point of view of statistical learning theory: it emphasizes one possible region of the weight space where generalization ability can be formally proved. It is well known that standard recurrent neural networks are not distribution independent learnable in the probably approximately correct (PAC) sense if arbitrary precision and inputs are considered. We prove that recurrent networks with contractive transition function with a fixed contraction parameter fulfill the so-called distribution independent uniform convergence of empirical distances property and hence, unlike general recurrent networks, are distribution independent PAC learnable.

Combination of Recurrent Neural Network and Deep Learning for Robot Navigation Task in Off-Road Environment

Robotica ◽  
2019 ◽  
Vol 38 (8) ◽  
pp. 1450-1462
Author(s):  
Farinaz Alamiyan-Harandi ◽  
Vali Derhami ◽  
Fatemeh Jamshidi
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Recurrent Neural Network ◽  
Robot Navigation ◽  
Recurrent Network ◽  
Navigation Task ◽  
Sensory Data ◽  
Uneven Terrain ◽  
Target Network ◽  
Fully Connected

SUMMARYThis paper tackles the challenge of the necessity of using the sequence of past environment states as the controller’s inputs in a vision-based robot navigation task. In this task, a robot has to follow a given trajectory without falling in pits and missing its balance in uneven terrain, when the only sensory input is the raw image captured by a camera. The robot should distinguish big pits from small holes to decide between avoiding and passing over. In non-Markov processes such as the abovementioned task, the decision is done using past sensory data to ensure admissible performance. Applying images as sensory inputs naturally causes the curse of dimensionality difficulty. On the other hand, using sequences of past images intensifies this difficulty. In this paper, a new framework called recurrent deep learning (RDL) with combination of deep learning (DL) and recurrent neural network is proposed to cope with the above challenge. At first, the proper features are extracted from the raw image using DL. Then, these represented features plus some expert-defined features are used as the inputs of a fully connected recurrent network (as target network) to generate command control of the robot. To evaluate the proposed RDL framework, some experiments are established on WEBOTS and MATLAB co-simulation platform. The simulation results demonstrate the proposed framework outperforms the conventional controller based on DL for the navigation task in the uneven terrains.

Sign in / Sign up

Export Citation Format

Share Document