Revolutionary AI inspired by sea slugs and octopuses learns to navigate and explore on its own to overcome obstacles: Tech : Tech Times

Researchers have developed revolutionary AI inspired by the navigation skills of a sea slug and the episodic memory of an octopus. This new approach uses basic associative learning rules augmented with improved episodic memory.

New AI with improved episodic memory, technology inspired by Octopus

Researchers at the University of Illinois created this AI that can navigate new environments, map landmarks, seek out rewards, and overcome obstacles. They built it based on their previous work simulating neural circuits similar to those in sea slugs.

Researchers developed this virtual creature called ASIMOV, inspired by Isaac Asimov, to reproduce decision-making processes. Interesting Engineering reported that ASIMOV was designed to monitor its current state, seek validation, and seek rewards in real time.

However, he struggled to learn from past experiences due to a lack of integrated retention of information. To address this limitation, researchers introduced the Feature Association Matrix, a computer module intended to improve episodic memory.

This module allows ASIMOV to encode and recall the spatial and temporal contexts of past events and experiences. According to the authors, episodic memory is a crucial element of natural intelligence, a component currently absent from most AI models.

The research team investigated brain network methodologies inspired by octopus behavior, naming their agent CyberOctopus within the ASIMOV-FAM framework.

The ASIMOV agent uses cognitive maps generated by the FAM to understand its environment, allowing it to create new paths and efficient shortcuts to navigate its environment to maximize rewards. This ability represents advanced spatial reasoning.

Additionally, they said this approach can extend beyond spatial navigation to improve efficiency and tackle more abstract, non-spatial tasks. They anticipated ASIMOV-FAM’s potential to be more computationally efficient and problem-solving with minimal training.

The team envisioned a future in which AI can learn autonomously and adaptively, much like how children learn, focusing on fundamental development that reduces reliance on large data and promotes creativity and adaptive behavior.

Also read: Advances in AI could end the need for animal testing in research

Pioneering AI with nature-inspired learning

The AI ​​is designed to help navigate unfamiliar environments, identify rewards, and create landmark maps while deftly overcoming obstacles. The team achieved this by incorporating simple learning rules similar to those used by sea slugs to forage, bolstered by advanced episodic memory reminiscent of octopuses.

The researchers are optimistic that this innovative approach will enable AI to efficiently explore and collect comprehensive spatial and temporal data, thereby improving its knowledge base through practical experience.

They noted that this method makes AI more similar to animals than current models, aiming to bridge the gap between basic memory functions seen in sea slugs and more complex human abilities.

The team believes this approach is significantly more efficient and generates more comprehensive data than existing methods. By integrating a memory module, AI can retain past information and potentially move from simple spatial learning to handling more complex cognitive tasks.

They also highlighted the versatility of these associative learning techniques, which could extend to tasks such as understanding sequences of motor behavior, mapping social networks, and solving linguistic problems.

Related article: Hong Kong scientists pioneer human brain-inspired AI advances for lifelong learning

ⓒ 2024 TECHTIMES.com All rights reserved. Do not reproduce without permission.