Alfonso Llanes, studied at Florida International University
Big data and analytics are today at the top of the corporate agenda. Together, they promise to transform the way companies do business, delivering the kind of performance gains last seen when computers became of age in the 1990s, and organizations redesigned their core processes. As data-driven strategies become part of the business playground, they will become an increasingly important point of competitive differentiation.
With big data companies are able to identify, combine, and manage multiple sources of inputs. This data gathering provides the capability to build advanced-analytics models for predicting and optimizing outcomes.
A clear vision of the desired business impact must shape the combined approach to data sourcing, model building, and organizational transformation. This will help avoid the common trap of asking what big data strategy can do.
Deep learning is a disruptive technology like the Internet and mobile computing once were. Open source software has been the dominant platform that has enabled these technologies to evolve. When a company strategy is to use disruptive technology or a disruptive business model, the results can be spectacular and can leave the competition far behind the end line. The reason for this developmental trend is that company’s growth might seem linear at first; it eventually exposes itself as being exponential. After this point, it becomes very difficult, if not impossible, for competitors to catch up.
Deep learning algorithms continue to mature at a rapid pace. In the beginning, frameworks exploited the available matrix multiplication libraries. These finely tuned algorithms have been developed over decades to its present state. As research continues, newer kinds of algorithms are being proposed.
Aside from being a very recent technology, the software that supports deep learning is a complex stack. A common perception is that most deep learning frameworks (i.e. TensorFlow, Torch, Caffe etc) are open source and many developers can participate the world over on making improvements to the knowledge and master many capabilities not yet discovered.
No comments:
Post a Comment