2019-321
2019-321
Anomaly Detection and Zero-Shot Learning with Error-Correcting Output Codes
Hieu Nguyen, SCOTT ZOCKOLL, Shen-Shyang Ho, MOHAMMED S. KHAN, LUCAS J. LAVALVA, and MATHEWR. MARCHIANO
Zero-shot learning (ZSL) is a process by which a machine learns to recognize previously unseen objects. ZSL is a crucial step in artificial intelligence as a machine has to self-learn new concepts without a human teacher continuously providing labeled training data to the machine. Many approaches have been proposed using transfer learning of seen objects or knowledge transfer of seen data to a semantic embedded space to learn the new concept. We propose a new direction for ZSL in a more realistic incremental learning setting such that when a machine recognizes an unseen object as an anomaly (a deviation from all seen objects), it encodes the object as a new concept and integrates the new concept to the existing predictive model. In particular, we take an error-correcting output code (ECOC) driven approach in machine learning where class labels are represented by codewords from an error-correcting code that, when arranged as a matrix, has large row and column Hamming distances. Using multilabel classification, anomalies are detected and encoded as new concepts based on the output codewords of the unseen objects and their Hamming distances with respect to seen objects. Preliminary results on a set of datasets show the feasibility of our proposed approach.
College of Science & Mathematics
Main Menu
- Computer Science
- Academic Programs
- BA Computing & Informatics
- BA Computer Systems Technology
- BS Computer Science
- Minor in Computer Science
- Minor in Cybersecurity
- Concentrations BS CS
- Concentrations BA C&I
- Certificates of Undergraduate Study
- MS Computer Science
- MS Cybersecurity
- Certificates of Graduate Study
- Accelerated Dual Degree Program
- Cybersecurity
- Data Science Programs
- Compare University Computing Programs
- Compare our Undergraduate Programs
- Advising Materials
- Undergraduate
- BS Computer Science
- BA Computing & Informatics
- BA Computer Systems Technology
- Certificate of Undergraduate Studies
- Computer Programming
- Mobile Apps CUGS
- Fundamental Computing CUGS
- Cybersecurity
- Blockchain Technologies and Cryptocurrencies
- Advanced Network Technology
- Azure Fundamentals
- Cybersecurity in Information Technology
- Database Development
- Database Fundamentals
- Digital Forensics
- Ethical Hacking
- Internet of Things
- Intrusion Detection/Prevention
- Linux Systems Administration
- Network Fundamentals
- Operating Systems Fundamentals
- Minor Degrees
- CS Undergraduate Catalog
- Graduate
- "4+1" (ADDP)
- Program Guides
- BA Computing & Informatics
- BA Computer Systems Technology
- BS Computer Science
- BS Data Science
- Minor in Computer Science
- Concentrations
- CUGS Guides
- Advanced Network Technology
- Azure Fundamentals
- Blockchain Technologies & Cryptocurrencies
- Computer Programming
- Cybersecurity
- Cybersecurity in Information Technology
- Database Development
- Database Fundamentals
- Digital Forensics
- Ethical Hacking
- Fundamental Computing
- Internet of Things
- Intrusion Detection/Prevention
- Linux Systems Administration
- Mobile Application Development
- Network Fundamentals
- Operating Systems Funamentals
- 4+1 Programs
- MS Computer Science
- MS Cybersecurity
- MS Data Science
- COGS Guides
- PhD in Data Science
- Standard Course Syllabi
- Forms & Policies
- Undergraduate
- Faculty and Staff
- Students
- Research
- News
- Events
- Contacts
- Faculty Portal - secured
- Site Index
- Can't find it?