Research & Development

Deep Learning with Neural Networks

Deep learning architectures are built using multiple levels of non-linear aggregators, for instance neural nets with many hidden layers. In this introductory talk Will Stanton discusses the motivations and principles regarding learning algorithms for deep architectures. Bill provides a primer to neural networks, and deep Learning. He explains how Deep Learning gives some of the best-ever solutions to problems in computer vision, speech recognition, and natural language processing.

and also, why Google is Investing in deep learning.



Finding Interesting Outliers - A Belief Network based Approach @ IEEE SoutheastCon 2015

Presented in the IEEE SoutheastCon 2015




Finding Interesting Outliers - A Belief Network based Approach


Abstract: Outliers are deviations from the usual trends of data; to discover interestingness among outliers i.e. finding anomalies which are of real-interest for subject matter experts is an active area of research in data mining and maching learning community. Due to its subjective nature, the definition of what amounts to ’interesting’ varies between domains and subject matter experts. In this research, we explore the quantification for measures of interestingness, using Bayesian Belief Networks as background knowledge. Mining outliers may help discover potential anomalies and fraudulent activities. Meaningful outliers can be retrieved and analyzed by using domain knowledge. Domain knowledge (or background knowledge) is represented using probabilistic graphical models such as Bayesian belief networks. Bayesian networks are graph-based representation used to model and encode mutual relationships between entities. Due to their probabilistic graphical nature, Belief Networks are an ideal way to capture the sensitivity, causal inference, uncertainty and background knowledge in real world data sets. Bayesian Networks effectively present the causal relationships between different entities (nodes) using conditional probability. This probabilistic relationship shows the degree of belief between entities. A quantitative measure which computes changes in this degree of belief acts as a sensitivity measure. In this research paper we provide an overview of interestingness measures, their use to measure sensitivity in belief networks and review the earlier work on so-called Interestingness Filtering Engine. Building upon these foundation, we introduce our algorithm IBOX - Interestingness based Bayesian Outlier eXplainer, which provides progressive improvement in the performance and sensitivity scoring of the earlier works. IBOX provides an iterative model to use multiple interestingness measures resulting in better performance and improved sensitivity analysis. The approach quantitatively validates probabilistic interestingness measures as an effective sensitivity analysis technique in rare class mining.

Topic Category: Data Mining and Machine Learning



Announcing NL-ESB - A Negative Latency Enterprise Service Bus

Download Paper - NL-ESB - A Negative Latency Enterprise Service Bus



Monads by David Crockford

The monadic curse is that once someone learns what monads are and how to use them, they lose the ability to explain it to other people.

Excellent lecture. Transcript and Monads for Humans


State of the IoT Security

In a recent podcast by Scott Hanselman and Erica Stanley, an Internet of Things (IoT) primer, the guest mentioned how security is being treated as an afterthought for most things IoT. This is unfortunately true in various areas of software development; but especially with the unprecedented growth of IoT, this lax in providing security standards will fast become a safety and security dilemma.

To borrow the variety, velocity and volume analogy of Big Data, IoT is also subject to a very large variety of devices, supporting different velocities (performance capacities) and volumes (large number of devices, meshes etc). Protection of data in these devices and providing privacy of is definitely the key challenges in the IoT. It is also bad for business since lax security measures will cause decreased adoption impacting the success of the IoT and hinder overall development.

Following are some of the relevant links and papers which provide overview, analysis and taxonomy of security and privacy challenges in IoT.


References and Further Reading


Penetration Testing techniques in Web Applications - Infographic

Penetration Testing techniques in web applications by Dimitris Mandilaras, Nikolaos Tsalis is an succinct info-graphic review of different security frameworks / methodologies including OWASP, PTES, ISSAF, NIST, OSSTM and PTF.

A short poster can be downloaded from here.



Selection of 2014 F# / Functional Programming Resources


Gradient Boosting Machine Learning by Prof. Hastie

Here is Prof. Hastie's recent talk from the H2O World conference. In this talk, professor Hastie takes us through Ensemble Learners like decision trees and random forests for classification problems.


Other excellent talks from the conference include the following.

  • Michael Marks - Values and Art of Scale in Business
  • Nachum Shacham of Paypal - R and ROI for Big Data
  • Hassan Namarvar, ShareThis - Conversion Estimation in Display Advertising
  • Ofer Mendelevitch, Hortonworks - Bayesian Networks with R and Hadoop,
  • Sandy Ryza, Cloudera - MLlib and Apache Spark
  • Josh Bloch, Lord of the APIs - A Brief, Opinionated History of the API
  • Macro and Micro Trends in Big Data, Hadoop and Open Source
  • Competitive Data Science Panel: Kaggle, KDD and data sports
  • Practical Data Science Panel

The complete playlist can be found here.


Teaching Functional Programming to Professional .NET Developers

An informative paper by Tomas Petricek of University of Cambridge.

Abstract. Functional programming is often taught at universities to first-year or second-year students and most of the teaching materials have been written for this audience. With the recent rise of functional programming in the industry, it becomes important to teach functional concepts to professional developers with deep knowledge of other paradigms, most importantly object-oriented. We present our experience with teaching functional programming and F# to experienced .NET developers through a book Real-World Functional Programming and commercially offered F# trainings. The most important novelty in our approach is the use of C# for relating functional F# with object-oriented C# and for introducing some of the functional concepts. By presenting principles such as immutability, higher-order functions and functional types from a different perspective, we are able to build on existing knowledge of professional developers. This contrasts with a common approach that asks students to forget everything they know about programming and think completely differently. We believe that our observations are relevant for trainings designed for practitioners, but perhaps also for students who explore functional relatively late in the curriculum.



Honorable mention to A Look at F# from C#’s corner 


Dissertation - Done!

Screen Shot 2014-08-26 at 10.37.48 AM

Go to Top