Sökning: "relative entropy"
Visar resultat 1 - 5 av 15 uppsatser innehållade orden relative entropy.
1. Uncertainty Quantification in Deep Learning for Breast Cancer Classification in Point-of-Care Ultrasound Imaging
Master-uppsats, Lunds universitet/Matematik LTHSammanfattning : Breast cancer is the most common type of cancer worldwide with an estimate of 2.3 million new cases in 2020, and the number one cause of cancer-related deaths in women. LÄS MER
2. Development of a QRNG front-end for shot noise measurement : analysis of quantum shot noise originating from photodiodes
Master-uppsats, Linköpings universitet/InformationskodningSammanfattning : As one of the more mature quantum technologies, quantum random number generators (QRNGs) fill an important role in producing secure and private keys for use in cryptogra- phy in e.g. quantum key distribution (QKD) systems. Many available QRNGs are expen- sive and optical QRNGs often require complex optical setups. LÄS MER
3. Investigating the influence of the tidal regime on harbour porpoise Phocoena phocoena distribution in Mount’s Bay, Cornwall
Master-uppsats, Lunds universitet/Institutionen för naturgeografi och ekosystemvetenskapSammanfattning : Investigating the influence of the tidal regime on harbour porpoise Phocoena phocoena distribution in Mount’s Bay, Cornwall Abstract Unintentional by-catch in fishing gear is a significant cause of mortality of harbour porpoises in UK waters. Understanding the spatial distribution of harbour porpoises at fine scales and how this changes over time is essential when trying to understand where these lethal interactions might occur. LÄS MER
4. Reinforcement learning for train dispatching : A study on the possibility to use reinforcement learning to optimize train ordering and minimize train delays in disrupted situations, inside the r ail simulator OSRD
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Train dispatching is a complex process, especially when the train traffic is disrupted, as the decisions taken by the dispatchers can have substantial consequences on the delays of the trains. The most frequent dispatching decisions consists in changing the order of trains at convergence points, where two tracks unite to become a single track. LÄS MER
5. Incorporating Metadata Into the Active Learning Cycle for 2D Object Detection
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : In the past years, Deep Convolutional Neural Networks have proven to be very useful for 2D Object Detection in many applications. These types of networks require large amounts of labeled data, which can be increasingly costly for companies deploying these detectors in practice if the data quality is lacking. LÄS MER