Research
[2024] Deep Reinforcement Learning for Multivariate Supply Chain Optimisation
An approach that leverages deep reinforcement learning (DRL) to tackle this challenge as a multi-objective and multi-variate supply-chain optimisation problem.
Authors: Franck Romuald Fotso Mtope
[ONGOING]
[2023] Multi-Aspectual Knowledge Elicitation for Procurement Optimization in a Warehouse Company
This paper introduces the Multi-Aspectual Knowledge Elicitation framework (MAKE4ML) — a novel approach designed to effortlessly and effectively extract valuable information from domain experts. This framework inherently facilitates the development of machine-learning models capable of optimizing business processes, thereby diminishing reliance on experts. The framework's application within a food warehouse company is showcased, specifically targeting the enhancement of the procurement process. The employed methodology revolves around conducting comprehensive interviews with procurement experts, thereby enabling a meticulous exploration of diverse facets inherent to a business process. Subsequently, the gathered insights are employed to conceive and calibrate a machine learning model (time series forecasting). This model effectively emulates the domain experts' proficiency, offering invaluable decision-oriented insights.
Authors: Franck Romuald Fotso Mtope, Sina Joneidy, Diptangshu Pandit, Farzad Rah
Paper accepted at CONVR 2023 (November 13-16, 2023) Florence, Italy
[2020] Region-based Deep Hashing for Multi-Instance Aware Image Retrieval
An instance-aware hashing approach Region-DH for large-scale multi-label image retrieval. The accurate object bounds can significantly increase the hashing performance of instance features. We design a unified deep neural network that simultaneously localizes and recognizes objects while learning the hash functions for binary codes. Region-DH focuses on recognizing objects and building compact binary codes that represent more foreground patterns. Region-DH can flexibly be used with existing deep neural networks or more complex object detectors for image hashing.
Authors: F.R. Fotso Mtope, B. Wei
Paper accepted at IJCNN 2020 (July 19-20, 2020) Glasgow, UK
[2020] Adaptive bone abnormality detection in medical imagery using deep neural networks
This research conducts transfer learning with optimal training option identification for the detection of wrist bone abnormalities in X-Ray imagery. Specifically, transfer learning based on Convolutional Neural Networks (CNNs), such as ResNet-18 and GoogLeNet, has been developed for wrist bone abnormality detection. The effect of altering the number of epochs on the network performance using an automatic process is also investigated.
Authors: Oliver Storey, Bo Wei, Li Zhang, Franck Romuald Fotso Mtope
Paper accepted at FLINS 2020 (August 18-21, 2020) Cologne, Germany