National Repository of Grey Literature 2 records found  Search took 0.01 seconds. 
Adversarial Examples in Machine Learning
Kocián, Matěj ; Pilát, Martin (advisor) ; Neruda, Roman (referee)
Deep neural networks have been recently achieving high accuracy on many important tasks, most notably image classification. However, these models are not robust to slightly perturbed inputs known as adversarial examples. These can severely decrease the accuracy and thus endanger systems where such machine learning models are employed. We present a review of adversarial examples literature. Then we propose new defenses against adversarial examples: a network combining RBF units with convolution, which we evaluate on MNIST and get better accuracy than with an adversarially trained CNN, and input space discretization, which we evaluate on MNIST and ImageNet and obtain promising results. Finally, we explore a way of generating adversarial perturbation without access to the input to be perturbed. 1
Metrics for eye movements comparisons
Kocián, Matěj ; Děchtěrenko, Filip (advisor) ; Vodrážka, Jindřich (referee)
Measurement of eye movements is becoming a well established part of expe- rimental research in many areas (such as human-computer interaction, cognitive psychology and others). Then usually a need arises to mutually compare the eye movements. Many different metrics have been suggested for this purpose, but what is missing is a comparison of these metrics and consequently an agreement on the ones that should be used in specific cases. In this thesis we describe some commonly used metrics and then create a model of smooth pursuit eye move- ments. We subsequently use this model to compare the ability of Levenshtein metric, Normalized Scanpath Saliency for dynamic scenes and discrete Fréchet distance to recognise similarity between the original eye movement trajectory and its modified copy. 1

See also: similar author names
2 Kocian, M.
2 Kocian, Martin
2 Kocián, M.
2 Kocián, Marek
2 Kocián, Martin
5 Kocián, Michal
3 Kocián, Miroslav
Interested in being notified about new results for this query?
Subscribe to the RSS feed.