DEEP LEARNING

Deep learning is the fastest growing field and the new big trend in machine learning. It revolutionizes the way we see Artificial Intelligence. A new area of Machine Learning research, which has been introduced with the objective of moving Machine Learning closer to one of its original goals of Artificial Intelligence. Deep Learning is a set of algorithms that use artificial neural networks to learn in multi-levels, corresponding to different levels of abstraction.

Deep Learning

Image: Creative Commons

Some of the applications of deep learning are automatic speech recognition, image recognition/Optical character recognition, NLP, and classification/clustering/prediction of almost any entity that can be sensed & digitized. Visual search is one of the many fields transformed in recent years by advances in deep learning (Pinterest introduced automatic object detection). With deep learning in use Business Intelligence Queries are now 100x faster. MapD’s database intelligently partitions, compresses and caches data across all GPUs. “Big Sur”, Facebook’s AI brain uses enormous amount of data, funneled in from all over the world to train AI processes that power board-game-playing programs and help software “read” photos and explain their content back to users. Lots of research work is happening around Deep Learning along with some practical solutions. The popular libraries are being used are Torch (by Facebook), Theano and TensorFlow (Mid 2016 Google moved to TensorFlow).

TensorFlow is an open source software library for numerical computation using data flow graphs. Nodes in the graph represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) communicated between them. The flexible architecture allows to deploy computation to one or more CPUs or GPUs in a desktop, server, or mobile device with a single API. TensorFlow isn’t a rigid neural networks library. If you can express your computation as a data flow graph, you can use TensorFlow. You construct the graph, and you write the inner loop that drives computation. New algorithms are being implemented and re-implemented in Tensorflow in a much speedier fashion than in the two mentioned libraries. We think that Tensorflow will move Torch from the almost dominant place it had, into a much more relegated place.

 

Leave a Reply

Your email address will not be published. Required fields are marked *