The expression computational intelligence (CI) usually refers to the ability of a computer to learn a specific task from data or experimental observation. Even though it is commonly considered a synonym of soft computing, there is still no commonly accepted definition of computational intelligence.  Generally, computational intelligence is a set of nature-inspired computational methodologies and approaches to address complex real-world problems to which mathematical or traditional modelling can be useless for a few reasons: the processes might be too complex for mathematical reasoning, it might contain some uncertainties during the process, or the process might simply be stochastic in nature. 

Pattern recognition is the automated recognition of patterns and regularities in data. It has applications in statistical data analysis, signal processing, image analysis, information retrieval, bioinformatics, data compression, computer graphics and machine learning.

Soft computing is a set of algorithms, including neural networks, fuzzy logic, and genetic algorithms. These algorithms are tolerant of imprecision, uncertainty, partial truth and approximation. It is contrasted with hard computing: algorithms which find provably correct and optimal solutions to problems. The theory and techniques related to soft computing were first introduced in 1980s.The term "soft computing" was coined by Lotfi A. Zadeh.

Granular computing (GrC) is an emerging computing paradigm of information processing that concerns the processing of complex information entities called "information granules", which arise in the process of data abstraction and derivation of knowledge from information or data. Generally speaking, information granules are collections of entities that usually originate at the numeric level and are arranged together due to their similarity, functional or physical adjacency, indistinguishability, coherency, or the like. Fort our research GrC is not only an algorithmic paradigm but also a way of multidimensional and systemic thinking for synthesizing intelligent procedures and models in complex systems (learn more).

The problem of representation is a core issue for pattern recognition. It encodes the real-world objects by some numerical description, handled by computers in such a way that the individual object representations can be interrelated. Based on that, later a generalization is achieved, establishing descriptions or discriminations between classes of objects. Originally, the issue of representation was almost neglected, as it was reduced to the demand of having good features provided by some experts. The learning is often believed to start at the given feature vector space. Indeed, many books on pattern recognition disregard the topic of representation, simply by assuming that objects are somehow already represented. (Learn More)

Businesses use machine learning to improve their products, services, and operations. By leveraging large amounts of historical data, businesses can build models to predict customer behaviors and refine internal processes. While machine learning provides incredible value to an enterprise, current CPU-based methods can add complexity and overhead reducing the return on investment for businesses.  With a data science acceleration platform that combines optimized hardware and software, the traditional complexities and inefficiencies of machine learning disappear. Data scientists can now conduct rapid feature iteration, use massive datasets for highly accurate predictions, and deliver value-generating solutions to production with ease. Data scientists can easily access GPU-acceleration through some of the most popular Python or Java-based APIs, making it easy to get started fast whether in the cloud or on-premise.  By leveraging the power of accelerated machine learning, businesses can empower data scientists with the tools they need to get the most out of their data. (Learn More)

Artificial Neural Network firing process

Information flow in CNNS