Identity Verification System
Location: Subiaco, WA
Duration: 5 months
Proposed start date: October 2019
Within Australia alone, identity fraud is responsible for close to $3 billion in losses annually in the financial and retail sectors and affects more than 1.6 million Australians. (https://www.abs.gov.au/ausstats/abs@.nsf/mf/4528.0)
Increasingly online platforms are faced with fake personas, bot accounts and phishing scams that are becoming difficult to distinguish in the complex environment of online commerce, social media misinformation and dark-web trade of personally identifiable information.
This type of fraud exploits the veracity of physical identity as our primary means of identity confirmation.
Meanwhile organisations that take steps to collect identification from their users are often working in isolation to others, duplicating efforts, implementing non-standard or ineffectual privacy and data collection controls, centred around the digital retention of personally identifiable information.
Scantek is developing a system to verify the identity of users with close to zero retention of personally identifiable information using a range of state-of-the-art techniques. Such a system would have benefits to everyday users who wish to present their ID and retain more granular control over when and when their personally identifiable information is made available. In addition, this solution could service businesses who need to properly verify a users ID and/or provide a mechanism to securely interchange ID with 3rd parties to provide services requiring authenticity
Research to be Conducted
By June 30th 2020, Scantek researchers are looking to engineer a commercially viable solution to the Open Set Classification problem – specifically being able to detect anomalous documents, including new classes for curation or fake documents with a super-human level of accuracy, with minimal human support. The road to achieve this requires that a number of key research objectives are met:
- Evolutionary development of deep learning architectures
- Evolutionary ensembling of datasets and model architectures
- Classification, Object Localisation and Segmentation
- Visual Question and Answering
- Advanced anomaly detection techniques.
- Advanced DevOps pipeline and support tools.
We are looking for a PhD student with the following:
- Computer Science preferably with interest in Computer Vision
- Mathematics and Statistics
- Machine Learning, Data Science and Data Engineering
- Tensorflow and/or TFX
- Unstructured Databases (NoSQL e.g. Mongo)
- Genetic Algorithms, Optimisation or Operations Research
- Kubernetes, Docker, Jupyter
By June 30th 2020, Scantek has a technology that is able to:
- Accelerate the identification of new ID types
- Verify their authenticity and validity,
- Generate new models for unknown classes in an intelligent pseudo-automatic process.
- Able to read any identification document and provide a measure of trust without retaining the document itself.
The intern will receive $3,000 per month of the internship, usually in the form of stipend payments.
It is expected that the intern will primarily undertake this research project during regular business hours, spending at least 80% of their time on-site with the industry partner. The intern will be expected to maintain contact with their academic mentor throughout the internship either through face-to-face or phone meetings as appropriate. Because of the nature of the datasets involved, access is strictly controlled and physical attendance on-site (Subiaco, Western Australia) is compulsory. Students will be required to complete an NDA, CA and pass police clearance checks. Training and support tools will be provided.
The intern and their academic mentor will have the opportunity to negotiate the project’s scope, milestones and timeline during the project planning stage.
4 September 2019
APR – 1142