Ai Kenya Show and Tell : Projects on medical imaging and social roboticsical imaging and social robotics

The 1st Show and Tell event in 2020 for Ai Kenya will feature a variety of speakers who will be showcasing different topics they work on, such as Social robotics and cancer detection.

Mbithe Nzomo will present on a project titled “Hello, Goodbye: Training Social Robots to Manage Interactions with Humans’’ Robot technology has become prevalent across many different industries. The most common types of robots in the past have been industrial ones used in settings such as factories. However, the demand for robots with a more social aspect is increasing. This is fuelled by the need for skilled labour for roles such as patient and elderly care, personal assistance, and teaching. The supply gap for such roles can be bridged using social robots designed to interact and communicate with people. It is important for such robots to interact with humans safely and in socially appropriate ways as per human norms. In this project, a deep learning model is built using image data that highlights body language cues. The model is then implemented on a robot platform, Pepper from SoftBank Robotics. Using the model, the robot is successfully able to evaluate a scene to decide what action to take when faced
with a human, including whether to start an interaction or end an ongoing one.

Stephen Obonyo will present on: Multitask Learning & Transfer Learning Application to Cancer Detection. Invasive ductal carcinoma (IDC), also known as infiltrating ductal carcinoma, is a type of cancer that begins growing in a milk duct and invades the fibrous or fatty tissue of the breast outside of the duct. IDC is the most common form of breast cancer, representing 80 percent of all breast cancer diagnoses. Multitask Learning (MTL) and Transfer Learning (TL) are two key Machine Learning (ML) approaches which have been widely adopted to improve model’s performance. In Deep Learning (DL) context, these two learning methods have contributed to competitive results in various areas of application even if the size of the dataset is relatively small. While MTL involves learning from a key task and other auxiliary tasks simultaneously and sharing signals among them, TL focuses on the transfer of knowledge from already existing solution within the same domain. In this project, Stephen presents MTL and TL based models and their application to Invasive Ductal Carcinoma (IDC) detection.

Speaker Bios
Stephen Obonyo is currently working as Assistant Lecturer at Strathmore University, Nairobi, Kenya. He holds a Bachelor’s degree in Computer Science (JKUAT) and a Master’s degree in Business Intelligence (Strathmore). His research interests include Transfer Learning , Multitask Learning and their applications, Computer Vision, Question Answering Systems and Intelligent Systems Design. He is also keen on projects pivoted around Artificial Intelligence for the Common Good.

Mbithe Nzomo, a Senior Technology Consultant at EY (Ernest & Young) , is enthusiastic about using technology to drive impact and solve some of humanity’s most pressing problems. She recently completed a master’s degree in Advanced Computer Science at the University of Manchester, where she specialised in AI and worked on a deep learning and human-robot interaction project.

Event Program
2.30pm – 3.00pm : Attendees check in
3.00 pm – 3.10 pm : Introduction keynote by Ai Kenya representative
3.10pm – 3.40pm : Hello, Goodbye: Training Social Robots to Manage Interactions with Humans : – Mbithe Nzomo
3.40pm – 4.10pm : Multitask Learning & Transfer Learning Application to Cancer Detection – Stephen Obonyo
:4.10pm – 5.00pm: Q n A
5.00pm – 5.30pm : Networking, event ends, attendees leave at their convenience

Show and tell event series by Ai Kenya is meant to provide a platform for community members and guest speakers to showcase their projects and consequently get feedback and collaborations from the community.