Summer 2017 -
Data Scientist at Pythian
Machine Learning, Deep Learning, Big Data Analytics, CloudRamp Services, Google Certified Professional Data Engineer
2016 - 2018
Carleton University Computer Science M.Sc. candidate
Reinforcement Learning, Learning Automata, Pattern Recognition, Adaptive Data Structures. My adviser is John Oommen
2015-2016
Graduate Assistant at the University of Calabar
Published my first book on "Programming in Java: A Quick Starter Guide"
ISBN: 978-007-277-2
2014
Complusory National Service to Nigeria
Instructor, Database Design and Development
2009-2013
Babcock University: BSc
Concentrations in Artificial Intelligence and Decision Support Systems
Bio. I am a Masters graduate student in the School of Computer Science at Carleton University with a research focus on Learning Systems (encompassing learning automata and reinforcement learning), Machine Learning, and Deep Learning. I am also a Data Scientist at Pythian, a big data analytics company headquartered in Ottawa. I am a Google Certified Professional Data Engineer.

My research interests are in the theoretical and practical aspects of machine learning, the principles and algorithms that affect learning and how they are applied to specialized domains in making predictions, especially where there are rapid shifts in observable features as well as how mathematical principles, statistics and probabilistic reasoning can be applied to aid better decision making. I am also interested in researching the frontiers involving the synthesis of deep learning and reinforcement learning (what is now known as deep reinforcement learning), and how they can improve the learning task of an agent interacting with a non-deterministic, non-stationary environment (where exact mathematical analysis are no longer feasible), especially when the states of the environment are only partially observable.


Research

Benchmarking Decoupled Neural Interfaces with Synthetic Gradients
Artificial neural networks have a close and stringent coupling between the modules of neurons in the network. To solve this problem, synthetic gradients (SG) with decoupled neural interfaces (DNI) (Jaderberg, et al 2016) are introduced as a viable alternative to the backpropagation algorithm. This paper performs a speed benchmark to compare the speed and accuracy capabilities of SG-DNI as opposed to a standard neural interface using multilayer perceptron MLP.
Bisong, Ekaba
Built to Last or Built Too Fast? Evaluating Prediction Models for Build Times
Accepted, Proceedings of the 2017 IEEE/ACM 14th International Conference on Mining Software Repositories (MSR 2017). Buenos Aires, Argentina. 20-21 May 2017. This research focuses on finding a balance between integrating often and keeping developers productive. We propose and analyze models that can predict the build time of a job. Such models can help developers to better manage their time and tasks.
Bisong, Ekaba, Tran, Eric, and Baysal, Olga




Thesis

[PDF] On Adaptively Designing Datastructures within Datastructures: A Learning Automata Hierarchical Scheme (In progress)

Ekaba Bisong,
M.Sc. Thesis, 2018




Pet Projects

Cereal Classifier: TensorFlow Android App
This project builds a cereal box image classifier, based on the Google InceptionV3 model with TensorFlow. Using a technique known as "transfer learning" we can use a pre-trained model such as the InceptionV3 model as a starting point for building a custom image classifier to classify our own images different from the original 1000 classes that the pre-trained model was trained to classify. The app is trained to classify the following cereals: apple cinnamon cheerios, berry kix, chocolate cheerios, cocoa puffs, honeycomb, kelloggs all bran, kelloggs corn pops, kelloggs frosted flakes, kelloggs rice krispies, multi grain cheerios, and sugar crisp.
Credit Card Fraud Detection using Extreme Gradient Boosting
Credit card fraud detection is the science and art of detecting unusual activity in credit transactions. A major challenge to credit fraud detection research is the availability of real world data due to privacy and legal concerns. In this project, we take the supervised approach to identifying fraud. We will build a function to learn a transaction map to identify fraud from a specific number of attributes. This attributes have already been transformed to their principal components so much cannot be done in terms of feature engineering.
Predicting Forest Cover Type using Extreme Gradient Boosting (xgboost)
This project develops a classification model using XGBoost to accurately classify and predict forest cover type from a set of cartographic features. The data is gotten from the UCI Machine Learning repository submitted by Jock A. Blackard. This dataset was particularly interesting to me because I previously worked with EcoDev Konsult, an environmental consultancy agency, to estimate Greenhouse Gas Emissions from Land Use, Land Use Change and Forestry (LU-LUCF) in Nigeria using GIS, remote sensing, machine learning and statistical methods.
show more
Built to Last or Built Too Fast?
Long build times can be an issue when integrations are frequent. This project is a cubist “minimum-viable-product” PoC model to predict the build time of a job.
Analysis of Agricultural Trends in Nigeria: A Shiny App
This application displays a barplot of several agricultural trends in Nigeria from 1961 to 2012. The collected data is from the World Bank Country Database.

Ekaba Bisong 2016-2018
Design adapted from Andrej Karpathy