Since 2007, I have worked on 100+ personal and academic software projects. I’m in the process of updating this site to showcase some more recent projects.
Bayesian GAN: Paper Review and Tutorial (2018)
The paper Bayesian GAN by Y. Saatchi and A.G. Wilson introduces a new formulation for GANs that applies Bayesian techniques, and outperforms some of the current state-of-the-art approaches. It is a true generalization in that we can recover the original GAN formulation given a specific choice of parameters, but authors also show that this formulation allows to sample from a family of generator and discriminators which avoids mode collapse. Posteriors on the parameters of the generator and the disciminator are defined mathematically as a marginalization over the noise inputs. An algorithm is presented to sample from the posterior distributions defining generators and discriminators, using SGHMC and MC. We will first explain the intuition behind the paper, describe the most important mathematical underpinnings and apply the algorithm to a new, simple problem. Finally, we extend the model on a new, unreleased dataset and show how it performs in comparison to other state of the art methods.
Convolutional Autoencoders for Image Manipulation (2019)
Convolutional Autoencoders have an extensive and successful record in applications such as representation learning, dimensionality reduction and learning data transformations. Recently, the concept of autoencoding has received even more attention with a rise of generative applications, which also make use of this extremely versatile concept. In this workshop, we will first introduce the concept and inner workings of autoencoders. We will then have a hands-on session with attendees implementing and training autoencoders on their own.
Easy File Sharing for Local Networks (2014)
- developed file sharing solution for local networks.
- designed for simple user experience, features Metro-style system overlay for easy access.
- supports sharing of files and directories, screen and clipboard content.
CLFlow – A Software-based Cluster for TensorFlow (2017)
- developed software to create software-managed computing clusters to parallelize heavy computing tasks on inexpensive hardware setups.
- software supports automatic job distribution, dataset synchronization, live training supervision and administration using central master node.
- deployed in research lab, significantly aided multiple research projects.
A Hidden-Markov Model Implementation for POS-Tagging (2016)
- .NET-based implementation of a Hidden Markov Model (HMM) part-of-speech (POS) tagger.
- designed for the English language.
- supports useful visualization features, with one example shown below.
Using Feedforward Neural Networks for Color Based Grape Detection
Precise yield estimation is a relevant challenge for agriculture. In this paper we apply simple feedforward neural networks (FFNN) on the problem of color-based grape detection in field images, achieving an average classification rate of 93% near realtime. Our evaluation shows a detailed comparison between our FFNN approach and SVMs handled by up to date SVM implementations, revealing FFNN can slightly outperform SVMs regarding computation time while obtaining competitive results. Furthermore our results are not only competitive with state of the art results of pixelwise color-based classification in the application field of precision farming but show also new contributions by investigating the influence of using different color models and evaluating our classifiers on different lighting conditions and grape varieties. We also present a specialized software framework with extensive options for customization to manipulate and process agricultural data.
A Closer Look at Restaurant Recommender Systems (2017)
- visual analysis of yelp’s restaurant dataset.
- predictive modeling of user ratings.
- employed baseline models based on sample averages and regression, matrix factorization, profile distance-based models, attribute-based regression models and ensembles.
A Simple .NET-based SOM Implementation (2015)
- basic implementation of self-organizing maps (SOMs) for unsupervised learning.
- library provides extensive options for configuration.
- various visualization features.
ANN.NET - A Neural Network Toolkit for .NET (2015)
- toolkit for training and simulating arbitrary feedforward neural networks in .NET.
- graphical visualizations tools, e.g. of network structure, training procedure and weights.
- extensive benchmarking and configuration options in user interface.