Research

Research Contents

Development of technology to extract and infer rule-based knowledge
from various multi-model deep learning models

Objective

By using a technique for extracting various abstract internal symbolic knowledge from a deep learning model through inductive learning, this research aims to convert abstract internal symbolic knowledge into a human-readable knowledge graph, and to develop symbolic (KBN) reflecting the degree of uncertainty about internal symbolic knowledge. For knowledge expansion, the research also aims to develop a technology that applies the method of learning rules between similar knowledge existing in the established knowledge base (KBH), reflecting certainty, to new knowledge.

  • Korea university, Konkuk university

    Technology to convert internal expressions into human-readable formats

  • KAIST

    Knowledge expression technology that expresses the knowledge extracted from the neural network model trained on noisy data differently depending on the degree of uncertainty

  • Sungkyunkwan university, Sogang university

    Technology to expand the knowledge base by deriving new symbolic knowledge through inductive learning

  • Wisenut

    Development of an extraction engine and interface that can integrate multimodal knowledge from neural network models (visual, auditory, language)