This is the first article in a series about most used Java libraries, frameworks and API's in big data projects. Java, one of the most broadly used programming languages in big data projects, owes part of its popularity to its extensive ecosystem. Programming in Java provides the access to this ecosystem that consists of several libraries, frameworks, and APIs. Within a series of articles I am going to briefly describe the most used Java libraries, frameworks, and APIs for big data projects. There are numerous third-party libraries for Java programming language.

Every day it seems we are hearing of new advances made by AIs thanks to Machine Learning, from improving healthcare to beating us at poker, it is often easy to forget that, behind every successful robot, there's a clever human. The swift pace of change we are seeing today is due to a concerted effort across industry and academia to find practical uses for the ever-growing amount of data we are generating and collecting. So, in this post I am going to highlight some of the current movers'n' shakers, whose breakthroughs in machine learning are proving to be fundamental to developing the digital tools and technologies making AI possible, from social networks to self-driving cars, to the industrial internet. Ng has just resigned from his post as chief data scientist at Chinese online giant Baidu. As well as that he is the founder of the online training resource Coursera and associate professor at Stanford University's computer science department.

There is much debate among scholars and practitioners about what data science is, and what it isn't. Does it deal only with big data? Is data science really that new? How is it different from statistics and analytics? One way to consider data science is as an evolutionary step in interdisciplinary fields like business analysis that incorporate computer science, modeling, statistics, analytics, and mathematics.

Deep Learning has been getting more and more traction. It focuses on one section of Machine Learning: Artificial Neural Networks. This article explains why Deep Learning is a game-changer in analytics, when to use Deep Learning, and how visual analytics allows business analysts to leverage the analytic models built by a (citizen) data scientist. Deep Learning is the modern buzzword for Artificial Neural Networks, one of many concepts in Machine Learning that is used to build analytics models. A neural network works similarly to a human brain.

Powerful and cost-effective HPC platforms promote data fusion, reduce training time, and enable ultra-scale real-time data analytics to power deep learning systems. In today's digital climate, organizations of every size and industry are both collecting and generating enormous amounts of data that can potentially be used to solve the world's greatest problems--from national security and fraud detection to scientific breakthroughs and technological advancement. However, traditional analysis techniques and practices are not capable of rapidly delivering automated, real-time insights from the rising data volumes to the point that artificial intelligence (AI) is becoming vital to harnessing the full understanding of scientific and business data. The evolution of Big Data is driving a major paradigm shift in the field of AI, which is increasing the need for high performance computing (HPC) technologies that can support high performance data analytics (HPDA). According to an IDC report, the HPDA server market is projected to grow at a 26% CAGR through 2020, including an additional $3.9 billion in revenue by 2018.

Whether you use Cortana, Siri, or Ok Google, your smartphone or computer's ability to recognize your voice and answer your questions is a function of deep learning. The software has been trained to recognize the myriad ways a word can be pronounced, then understand the various types of questions that might be asked, and to provide an appropriate answer to meet your needs. Using data about when your home needs light, temperature adjustments, and more, smart home technology allows customers to reduce energy usage and save money. When combined with voice recognition tools, smart homes can have profound applications for disabled consumers. If you take pictures on your smartphone and store them in various cloud applications, you may have already encountered apps that will sort and organize albums based on various features.

H2O.ai, the company bringing AI to enterprises, today announced it has partnered with Nimbix to offer its next-generation AI platform, GPU-powered machine learning and best-of-breed Deep Learning on the Nimbix Cloud. With this partnership, Nimbix customers can launch H2O clusters and provision the entire stack in a matter of minutes using the latest GPUs in the Nimbix cloud. Enterprises can use this end-to-end solution to operate on large datasets, iterate faster, deploy quickly and gain real-time insights. H2O--s AI solutions enable customers to train machine learning and deep learning models up to 35x faster compared to conventional CPU-based solutions. The entire H2O platform can be easily provisioned and scaled on Nimbix--s state of the art infrastructure with JARVICE--, the cloud platform for deep learning and big compute.

Convolutional Neural Networks(CNN) has had a great success in the recent past, because of the advent of faster GPUs and memory access. CNNs are really powerful as they learn the features from data in layers such that they exhibit the structure of the V-1 features of the human brain. A huge bottleneck in this case is that CNNs are very large and have a very high memory footprint, and hence they cannot be employed on devices with limited storage such as mobile phone, IoT etc. In this work we study the model complexity versus accuracy trade-off on MNSIT dataset, and give a concrete framework of handling such a problem, given the worst case accuracy that a system can tolerate. In our work we reduce the model complexity by 236 times, and memory footprint by 19.5 times compared to base model, while achieving worst case accuracy threshold. The experiment on speed-accuracy trade-off for digit recognition using MNIST dataset [1] is done using tensorflow [2].

We present a new deep meta reinforcement learner, which we call Deep Episodic Value Iteration (DEVI). DEVI uses a deep neural network to learn a similarity metric for a non-parametric model-based reinforcement learning algorithm. Our model is trained end-to-end via back-propagation. Despite being trained using the model-free Q-learning objective, we show that DEVI's model-based internal structure provides `one-shot' transfer to changes in reward and transition structure, even for tasks with very high-dimensional state spaces.

Large datasets often have unreliable labels-such as those obtained from Amazon's Mechanical Turk or social media platforms-and classifiers trained on mislabeled datasets often exhibit poor performance. We present a simple, effective technique for accounting for label noise when training deep neural networks. We augment a standard deep network with a softmax layer that models the label noise statistics. Then, we train the deep network and noise model jointly via end-to-end stochastic gradient descent on the (perhaps mislabeled) dataset. The augmented model is overdetermined, so in order to encourage the learning of a non-trivial noise model, we apply dropout regularization to the weights of the noise model during training. Numerical experiments on noisy versions of the CIFAR-10 and MNIST datasets show that the proposed dropout technique outperforms state-of-the-art methods.