In this work, we review Binarized Neural Networks (BNNs). BNNs are deep neural networks that use binary values for activations and weights, instead of full precision values. With binary values, BNNs can execute computations using bitwise operations, which reduces execution time. Model sizes of BNNs are much smaller than their full precision counterparts. While the accuracy of a BNN model is generally less than full precision models, BNNs have been closing accuracy gap and are becoming more accurate on larger datasets like ImageNet. BNNs are also good candidates for deep learning implementations on FPGAs and ASICs due to their bitwise efficiency. We give a tutorial of the general BNN methodology and review various contributions, implementations and applications of BNNs.

For many real world problems we must perform classification under widely varying amounts of computational resources. For example, if asked to classify an instance taken from a bursty stream, we may have from milliseconds to minutes to return a class prediction. For such problems an anytime algorithm may be especially useful.In this work we show how we can convert the ubiquitous nearest neighbor classifier into an anytime algorithm that can produce an instant classification, or if given the luxury of additional time, can utilize the extra time to increase classification accuracy. We demonstrate the utility of our approach with a comprehensive set of experiments on data from diverse domains.

This study investigated the drop-spreading dynamics of pseudo-plastic and dilatant fluids. Experimental results indicated that the spreading law for both fluids is related to rheological characteristics or power exponent n. For the completely wetting system, the evolution of the wetting radius over time can be expressed by the power law R = atm, where the spreading exponent m of the dilatant fluids is >0.1 and the spreading exponent m of pseudo-plastic fluids is <0.1. The strength of non-Newtonian effects is positively correlated to the extent of deviation from the theoretical value 0.1 of m for Newtonian fluids. For the partially wetting system, the power law on the time dependence of the wetting radius no longer holds; therefore, an exponential power law, R = Req(1-exp(-at(m)/Req)), is proposed, where Req denotes the equilibrium radius of drop and a is a coefficient. Comparing experimental data with the exponential power law revealed that both are in good agreement.

No abstract