Low-Power Gesture Recognition Using Convolutional Neural Networks and Ambient Lighting

This paper presents a study focused on developing an efficient signal processing pipeline and identifying suitable machine learning models for real-time gesture recognition using a testbed consisting of an Arduino Nano 33 BLE and three OPT101 photodiodes. Our research aims to address the challenges of limited computational power whilst...

Recognising Gestures Using Ambient Light and Convolutional Neural Networks: Adapting Convolutional Neural Networks for Gesture Recognition on Resource-constrained Microcontrollers

This paper presents how a convolutional neural network can be constructed in order to recognise gestures using photodiodes and ambient light. A number of candidates are presented and evaluated, with the most performant being adopted for in-depth analysis. This network is then compressed in order to be ran on an Arduino Nano 33 BLE...

Using Explainable Artificial Intelligence to Improve Transparency of Reinforcement Learning for Online Adaptive Flight Control: Breaking Open the Black Box

Deep Reinforcement Learning (DRL) shows great potential for flight control, due to its adaptability, fault-tolerance, and as it does not require an accurate system model. However, these techniques, like many machine learning applications, are considered black-box as their inner workings are hidden. This paper aims to break open the black box of...

Deep Just-in-Time Defect Prediction at Adyen

Finding defects in proposed changes is one of the biggest motivations and expected outcomes of code review, but does not result as often as expected in actually finding defects. Just-in-time (JIT) defect prediction focuses on predicting bug-introducing changes, which can help with efficient allocation of inspection time according to the defect...

Detecting Rhyming Words

Rhyming words are one of the most important features in poems. They add rhythm to a poem, and poets use this literary device to portray emotion and meaning to their readers. Thus, detecting rhyming words will aid in adding emotions and enhancing readability when generating poems. Previous studies have been done on the topic of poem generation....

Design of a Graph Neural Network: to predict the optimal resolution of the Sonar Performance Model

Graph Neural Networks are a unique type of Deep Learning models that have a capability to exploit an explicitly stated structure of data representation. By design they carry a strong relational inductive bias, which is a set of assumptions that makes the algorithm prioritize some solutions over another, independent of observed data. This makes...

Investigating Physics-Informed Neural Networks for solving PDEs

Physics-Informed Neural Networks (PINNs) are a new class of numerical methods for solving partial differential equations (PDEs) that have been very promising. In this paper, four different implementations will be tested and compared. These include: the original PINN functional with equal weights for the interior and boundary loss, the same...

Applying Machine Learning to Learn System Dynamics Models for Urban Systems

System Dynamics (SD) is an approach to study the nonlinear behaviour of complex systems over time. SD models provide a high­level understanding of the system and aid in designing policies to achieve specific system behaviours. Conventional SD modelling requires an intensive amount of time, human resources and effort. Applying Machine Learning ...

Application of Deep Learning to Coherent Fourier Scatterometry data

This thesis discusses the application of deep learning to Coherent Fourier<br/>Scatterometry data in order to quickly and reliably detect nanoparticles on<br/>surfaces. An introduction to deep learning is followed by a review of the<br/>experimental setup and used software. After that, results are presented of<br/>classification accuracy tests...

Towards deeper understanding of semi-supervised learning with variational autoencoders

Recently, deep generative models have been shown to achieve state-of-the-art performance on semi-supervised learning tasks. In particular, variational autoencoders have been adopted to use labeled data, which allowed the development of SSL models with the usage of deep neural networks. However, some of these models rely on ad-hoc loss additions...

Black Magic in Deep Learning: Understanding the role of humans in hyperparameter optimization

Deep learning is proving to be a useful tool in solving problems from various domains. Despite a rich research activity leading to numerous interesting deep learning models, recent large scale studies have shown that with hyperparameter optimization it is hard to distinguish these models based on their final performance. Hyperparameter...

Deep Reinforcement Learning for Goal-directed Visual Navigation

Safe navigation in a cluttered environment is a key capability for the autonomous operation of Micro Aerial Vehicles (MAVs). This work explores a (deep) Reinforcement Learning (RL) based approach for monocular vision based obstacle avoidance and goal directed navigation for MAVs in cluttered environments. We investigated this problem in the...

Learning the scale of image features in Convolutional Neural Networks

The millions of filter weights in Convolutional Neural Networks (CNNs), all have a well-defined and analytical expression for the partial derivative to the loss function. Therefor these weights can be learned from data with a technique called gradient descent optimization. While the filter weights have a well-defined derivative, the filter size...

Tuning of Optical Beamforming Networks: A Deep Learning Approach

Optical beamforming networks (OBFNs), which consist of many small and flat antennas, called phased array antennas (PAAs), can be tuned such that the signal beam from the airplanes can be steered towards a satellite. This was proposed as a alternative to the mechanically steered antenna, which has many disadvantages. The problem of tuning a large...