Sökning: "manipulation och träning"
Visar resultat 1 - 5 av 8 uppsatser innehållade orden manipulation och träning.
1. Comparative Analysis of Spatiotemporal Playback Manipulation : Evaluating Desktop Environments versus Immersive Head-Mounted Virtual Reality Environments
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Virtual Reality (VR) is a creative tool that enables immersive learning, planning and training of surgical operations. Extensive research has been conducted in multiple surgical specialities where VR has been utilized, such as spinal neurosurgery. However, cranial neurosurgery remains relatively unexplored in this regard. LÄS MER
2. Copy-move Image Forgery Detection with Convolutional Neural Networks
Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Copy-move manipulation is a forgery method used on images where a small part is copied to another part. This thesis analyses the detection of copy-move forgeries with the help of Convolutional Neural Networks (CNN). The model used is utilizing an existing custom CNN layer to identify features useful for detecting manipulations. LÄS MER
3. Rehabiliteringsmetoder och samarbete vid ospecificerad ryggsmärta hos häst
Kandidat-uppsats, SLU/Dept. of Clinical SciencesSammanfattning : Ryggsmärta är vanligt förekommande hos hästar och ett flertal rehabiliteringsmetoder kan användas vid behandling av ryggsmärta. Hästarna rehabiliteras inte enbart av veterinärer utan olika yrkeskategorier arbetar med rehabilitering av häst. LÄS MER
4. Investigation of Different Observation and Action Spaces for Reinforcement Learning on Reaching Tasks
Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)Sammanfattning : Deep reinforcement learning has been shown to be a potential alternative to a traditional controller for robotic manipulation tasks. Most of modern deep reinforcement learning methods that are used on robotic control mostly fall in the so-called model-free paradigm. LÄS MER
5. Reinforcement learning for robotic manipulation
Master-uppsats, KTH/Skolan för datavetenskap och kommunikation (CSC)Sammanfattning : Reinforcement learning was recently successfully used for real-world robotic manipulation tasks, without the need for human demonstration, usinga normalized advantage function-algorithm (NAF). Limitations on the shape of the advantage function however poses doubts to what kind of policies can be learned using this method. LÄS MER