InsightInsight
sfi
  • About
    • Who We Are
    • What We Do
    • Our Structure
  • People
    • Work With Us
    • Senior Leadership
    • Principal Investigators
    • Funded Investigators
    • Research and Operations
  • Research
    • Excellence
    • Ecosystem
    • Funding Collaboration
    • National Projects
    • European Projects
    • Publications
  • Industry
    • Engage
    • Collaborate
    • Contact
  • Public Engagement
    • Highlights
    • What We Do
    • Meet the Team
    • Our Strategy
  • News
    • Latest News
    • Spotlight on Research
    • Newsletter
    • Press Releases
    • Media Queries
  • Contact
  • About
    • Who We Are
    • What We Do
    • Our Structure
  • People
    • Work With Us
    • Senior Leadership
    • Principal Investigators
    • Funded Investigators
    • Research and Operations
  • Research
    • Excellence
    • Ecosystem
    • Funding Collaboration
    • National Projects
    • European Projects
    • Publications
  • Industry
    • Engage
    • Collaborate
    • Contact
  • Public Engagement
    • Highlights
    • What We Do
    • Meet the Team
    • Our Strategy
  • News
    • Latest News
    • Spotlight on Research
    • Newsletter
    • Press Releases
    • Media Queries
  • Contact

Learning and Injecting Knowledge into Deep Neural Networks for more Robust and Effective Learning of new Task with limited Training Data

Insight>Projects>Learning and Injecting Knowledge into Deep Neural Networks for more Robust and Effective Learning of new Task with limited Training Data

Funding Programme or Company Name:

Govt. of Ireland Postgraduate Scholarship

Funding Body:

Irish Research Council (IRC)

Description:

Artificial Intelligence (AI) today enables machines to perform complex tasks by simulating the way human brain works. This relies on recent advances in Artificial Neural Networks, the most widely algorithm for AI that succeeded where AI failed for decades. Artificial Neural Networks today can leverage large amounts of available data to self-train themselves to perform hard tasks such as recognising objects in an image or understanding languages. Despite this recent success of Artificial Neural Networks, the process they use to perform these tasks cannot be interpreted or understood by humans. Therefore we cannot know how the system arrived at its conclusion or which steps it used to solve the task; all we know is how good the solution is. This research aims to help make the AI process based on Artificial Neural Networks understandable by humans. We also aim to incorporate two aspects the algorithm currently lacks: the ability to extract relationships such as causality between elements involved in learning to perform a task, and the ability to use background knowledge when learning, both typical of human decision making and reasoning. Decision due March 2018

Insight Contact:

Alessandra Mileo

Application Domain:

The Discovery Economy

Research Group:

Media Analytics

Associated Theme for Application Domain:

The Discovery Economy-Discovery Analytics

Involved Institution:

Dublin City University (DCU)

Non-Insight Contact:

Sean Quinn

footer-top
  • Privacy Statement
  • Copyright Statement
  • Data Protection Notice