A Computationally Inexpensive Classifier Merging Cellular Automata and MCP-Neurons

Abstract

There is an increasing need for personalised and context-aware services in our everyday lives and we rely on mobile and wearable devices to provide such services. Context-aware applications often make use of machine-learning algorithms, but many of these are too complex or resource-consuming for implementation on some devices that are common in pervasive and mobile computing. The algorithm presented in this paper, named CAMP, has been developed to obtain a classifier that is suitable for resource-constrained devices such as FPGA:s, ASIC:s or microcontrollers. The algorithm uses a combination of the McCulloch-Pitts neuron model and Cellular Automata in order to produce a computationally inexpensive classifier with a small memory footprint. The algorithm consists of a sparse binary neural network where neurons are updated using a Cellular Automata rule as the activation function. Output of the classifier is depending on the selected rule and the interconnections between the neurons. Since solving the input-output mapping mathematically can not be performed using traditional optimization algorithms, the classifier is trained using a genetic algorithm. The results of the study show that CAMP, despite its minimalistic structure, has a comparable accuracy to that of more advanced algorithms for the datasets tested containing few classes, while performing poorly on the datasets with a higher amount of classes. CAMP could thus be a viable choice for solving classification problems in environments with extreme demands on low resource consumption.