Introducing ASTRID /Brain 5
Introducing ASTRID /Brain 5

ASTRID, short for Analysis of Systemic Tagging Results in Intelligent Dynamics, is our premier technology and has been in development for over a decade. Welcome to the fifth generation of the ASTRID Cognitive Architecture.

Currently, at /BRAIN 5 status, the system boasts impressive feats in the AGI domain, thus far believed to be decades away from existence. Capable of Fully Autonomous Incremental Learning and building strong inferences across multiple knowledge domains, the ASTRID system can leverage its capacity for Cross Domain Deep Inference to find insights and answers to complex problems and situations.

ASTRID /BRAIN 5 has, through unsupervised learning, reached the equivalent conceptual knowledge of a Human Brain. Currently, at approximately 50.000 concepts and with close to 400.000 direct conceptual links, the ASTRID system can find an estimated more than 840.000.000 inference paths (calculated with 5 levels of inference depth).

Most impressive about the ASTRID system is the fact that the numbers stated here are accomplished by the system while running on very lightweight hardware; at our facilities ASTRID's benchmark installation runs on a (virtualized) 4-core CPU system with 8 Gigabytes of RAM. Running on this minimal hardware, ASTRID learns approximately 2-3 times as fast as humans. Obviously, when scaling up the hardware, ASTRID's processing capacity is virtually unlimited.




  • Brain 1: The Semantic Bootstrapping Trainer demonstrates capability to learn from 'Sparse Datasets'.
  • Brain 2: The ASTRID System builds its complex world model from curated training sentences.
  • Brain 3: ASTRID is capable of reading non-curated texts for full unsupervised learning.
  • Brain 4: The Emotional Bias Response Model ™ (EBRM) is implemented, putting emotion in the machine.
  • Brain 5: ASTRID gains the basic infrastructure to recognize different real-world instances of the same concept.
  • Brain 6: UPCOMING! Implementation of our Deep Inference Engine is on its way, currently undergoing rigorous development and testing.
  Internal Articles
  • ASTRID Fact Sheet - Dutch version 
  • ASTRID Fact Sheet - English version 
  Internal Papers & Reports
  • ASTRID: Bootstrapping Commonsense Knowledge - Hans Peter Willems (2021) 

© 2021-2023 MIND|CONSTRUCT  
Real  Emotion
Real Emotion

Within the ASTRID system, emotion is the key ingredient for real understanding. Without emotion there is no experience and without experience there is no understanding.

Before ASTRID, emotion has never been implemented in a machine at the fundamental level of cognitive capacities. Until now, emotion has only been used as a gimmick to show affect in a robot, mainly based on database lookups of certain facial expressions in humans. In stark contrast the ASTRID system implements our Emotional Bias Response Model ™ (EBRM) which basically gives 'feelings' to the machine.


By implementing emotion in the machine, and in concert with its internal complex world model, the ASTRID system automatically gains human-like abilities like empathy and moral insights. This guarantees the safe application of the ASTRID system in any domain, as it is capable of understanding and evaluating the moral consequences of possible actions.

Because of the virtually unlimited capability of its Cross Domain Deep Inference, ASTRID can even find moral consequences that are invisible to humans.

You can read more about the importance of Emotion in Cognitive Architectures.

© 2021-2023 MIND|CONSTRUCT  
ASTRID /Brain Specifications

Current Generation Brain 5
Product Class Cognitive Architecture
Concepts learned 50.000 (Human Brain equivalent)
Training capability Unsupervised Incremental Learning across Knowledge Domains
Human Equivalent Learning Performance 200% - 300% (on stated minimal hardware, benchmark development installation)
Inference Capability Deep Inference across Knowledge Domains
Minimal Hardware Platform Quad-core CPU running at 3 GHz, 8 Gigabyte RAM, 512 Gigabyte Storage
Development Platform Codedness™ IDE, PHP (proof-of-concept), Clojure (production)
Default System Communication Protocol Simple Text Oriented Message Protocol (STOMP)

© 2021-2023 MIND|CONSTRUCT