This blog post is intended for readers who have purchased a copy of my new book, Deep Learning for Computer Vision with Python. Inside this tutorial you'll learn how to: If you have any other questions related to the book, please send me an email or use the contact form. Thank you for picking up a copy of Deep Learning for Computer Vision with Python! I appreciate your support of both myself and the PyImageSearch blog. Without you, PyImageSearch would not be possible.

Momentum based stochastic gradient methods such as heavy ball (HB) and Nesterov's accelerated gradient descent (NAG) method are widely used in practice for training deep networks and other supervised learning models, as they often provide significant improvements over stochastic gradient descent (SGD). Rigorously speaking, "fast gradient" methods have provable improvements over gradient descent only for the deterministic case, where the gradients are exact. In the stochastic case, the popular explanations for their wide applicability is that when these fast gradient methods are applied in the stochastic case, they partially mimic their exact gradient counterparts, resulting in some practical gain. This work provides a counterpoint to this belief by proving that there exist simple problem instances where these methods cannot outperform SGD despite the best setting of its parameters. These negative problem instances are, in an informal sense, generic; they do not look like carefully constructed pathological instances. These results suggest (along with empirical evidence) that HB or NAG's practical performance gains are a by-product of mini-batching. Furthermore, this work provides a viable (and provable) alternative, which, on the same set of problem instances, significantly improves over HB, NAG, and SGD's performance. This algorithm, referred to as Accelerated Stochastic Gradient Descent (ASGD), is a simple to implement stochastic algorithm, based on a relatively less popular variant of Nesterov's Acceleration. Extensive empirical results in this paper show that ASGD has performance gains over HB, NAG, and SGD.

Distance metric learning is a successful way to enhance the performance of the nearest neighbor classifier. In most cases, however, the distribution of data does not obey a regular form and may change in different parts of the feature space. Regarding that, this paper proposes a novel local distance metric learning method, namely Local Mahalanobis Distance Learning (LMDL), in order to enhance the performance of the nearest neighbor classifier. LMDL considers the neighborhood influence and learns multiple distance metrics for a reduced set of input samples. The reduced set is called as prototypes which try to preserve local discriminative information as much as possible. The proposed LMDL can be kernelized very easily, which is significantly desirable in the case of highly nonlinear data. The quality as well as the efficiency of the proposed method assesses through a set of different experiments on various datasets and the obtained results show that LDML as well as the kernelized version is superior to the other related state-of-the-art methods.

The practical impact of deep learning on complex supervised learning problems has been significant, so much so that almost every Artificial Intelligence problem, or at least a portion thereof, has been somehow recast as a deep learning problem. The applications appeal is significant, but this appeal is increasingly challenged by what some call the challenge of explainability, or more generally the more traditional challenge of debuggability: if the outcomes of a deep learning process produce unexpected results (e.g., less than expected performance of a classifier), then there is little available in the way of theories or tools to help investigate the potential causes of such unexpected behavior, especially when this behavior could impact people's lives. We describe a preliminary framework to help address this issue, which we call "deep visual explanation" (DVE). "Deep," because it is the development and performance of deep neural network models that we want to understand. "Visual," because we believe that the most rapid insight into a complex multi-dimensional model is provided by appropriate visualization techniques, and "Explanation," because in the spectrum from instrumentation by inserting print statements to the abductive inference of explanatory hypotheses, we believe that the key to understanding deep learning relies on the identification and exposure of hypotheses about the performance behavior of a learned deep model. In the exposition of our preliminary framework, we use relatively straightforward image classification examples and a variety of choices on initial configuration of a deep model building scenario. By careful but not complicated instrumentation, we expose classification outcomes of deep models using visualization, and also show initial results for one potential application of interpretability.

A fresh wave of artificial intelligence rolling through Microsoft's language translation technologies is bringing more accurate speech recognition to more of the world's languages and higher quality machine-powered translations to all 60 languages supported by Microsoft's translation technologies. The advances were announced at Microsoft Tech Summit Sydney in Australia on November 16. "We've got a complex machine, and we're innovating on all fronts," said Olivier Fontana, the director of product strategy for Microsoft Translator, a platform for text and speech translation services. As the wave spreads, he added, these machine translation tools are allowing more people to grow businesses, build relationships and experience different cultures. Microsoft's research labs around the world are also building on top of these technologies to help people learn how to speak new languages, including a language learning application for non-native speakers of Chinese that also was announced at this week's tech summit. The new Microsoft Translator advances build on last year's switch to deep neural network-powered machine translations, which offer more fluent, human-sounding translations than the predecessor technology known as statistical machine translation.

Every year a large amount of data is generated which needs to be stored and analyzed. Apache Spark allows you to process such big data. The real power and value proposition of Apache Spark is its speed and platform to execute data science tasks. Spark's unique use case is that it combines ETL, batch analytic, real-time stream analysis, machine learning, graph processing, and visualizations to allow data scientists to tackle the complexities that come with raw unstructured data sets. Spark embraces this approach and has the vision to make the transition from working on a single machine to working on a cluster, something that makes data science tasks a lot more agile.

Suggestic is a personalized nutrition service that helps people make optimal food choices for weight loss, chronic disease reversal and health improvement. A: We start by creating a "precision eating" plan according to the user's personal goals and preferences, like vegetarian, gluten-free, or vegan. These plans are based on existing evidence-based protocols, for example, the Ketogenic diet, the Low FODMAP diet or the Mediterranean diet. Some of these plans are free, and others--like the ones curated by best-selling authors--are part of our subscription plan, Suggestic Plus. Once users have their personalized diets the app helps them make the best fitting food decisions.

Python is gaining an ever larger footprint in the interesting world of Data Science. KDnuggets says that Python R is even over as leader in the AI and Machine Learning standings platforms. But what makes Python so special for data science? One of the reasons is that Python is a general-purpose programming language. This means that Python has no specific purpose, which makes it possible to use models directly in a broader context.

As the world of technology progresses, Artificial Intelligence becomes more intertwined into every aspect of our lives. From marketing programs that collect and process huge amounts of data, smart fridges that know when we need to get more milk and the prototype self driving trucks slowly making their ways in our streets... Even the fields of teaching have accommodated AI with Intelligent Tutoring Systems such as the United States Air force's Own "Sherlock" program. Wikipedia defines AI ( Artificial Intelligence) as: " intelligence exhibited by machines, has many applications in today's society. More specifically, it is Weak AI, the form of A.I. where programs are developed to perform specific tasks, that is being utilized for a wide range of activities including medical diagnosis, electronic trading, robot control, and remote sensing. AI has been used to develop and advance numerous fields and industries, including finance, healthcare, education, transportation, and more.."

Since the early industrial age, we've been fascinated by self-operating devices. They represent the humanization of technology. Today, it is software that that's becoming more human -- most obviously "chatbots." But how do these machines work? First, wind back time and explore an earlier -- yet similar -- technology.