Yeshiva University
Data scientists have been able to leverage better algorithms on faster hardware optimized with graphical processing units to deliver improved performance and accuracy in whole classes of applications that had been previously commercially unviable. The biggest beneficiaries are applications that require unstructured data, such as audio and or video processing. Deep neural networks have also provided gains for other complex applications, from recommendation systems to natural language processing. This course builds on the concepts in machine learning to train multi-layered neural networks. Main topics covered in this course are generalization, convolutional neural network, recurrent neural network, long short-term memory, and autoencoder.
In classical programming, answers are obtained from rules and data. In machine learning, rules are obtained from data and answers. The widespread availability and sharing of data, and improvements in computing capacity, processing methods, and algorithms have given machine learning the power to deliver game-changing systems and technologies to organizations that compete on predictive, prescriptive, and/or autonomous analytics. In this course, we’ll look at methods for using, tuning, and comparing machine learning algorithms, based on measures of performance, accuracy, and explainability. We’ll also look at recent advances and trends in automated machine learning.
Stevens Institute of Technology
An introduction course for machine learning theory, algorithms and applications. This course aims to provide students with the knowledge in understanding key elements of how to design algorithms/systems that automatically learn, improve and accumulate knowledge with experience. Topics covered in this course include decision tree learning, neural networks, Bayesian learning, reinforcement learning, ensembling multiple learning algorithms, and various application problems. The students will have chances to simulate their algorithms in a programming language and apply them to solve real-world problems.
History of network security; classical infosec; cryptosecurity; Kerberos for IP networks; private and public keys; nature of network security; fundamental framework for network security; security on demand in ATM networks; analysis and performance impact of ATM network topology; security in IVCC; vulnerabilities and security attack models in ATM, IP and mobile wireless networks; intrusion detection techniques - centralized and distributed; emulation of attack models and performance assessment through behavior modeling and asynchronous distributed simulation; principles of secure network design in the future; projects in network security and invited guest lecturers.
The focus of the course is on data networks and end-user software environments for information systems. Topics include the TCP/IP protocols, organization of large-scale data networks, end-to-end operation over heterogeneous networks and the software foundation of client-server application programs. The students complete a project using TCP/IP protocols to create a basic client-server application.
University of Arkansas at Little Rock