Recursion and evolution: Part II

06/12/2020
by   Alexandros Arvanitakis, et al.
0

We examine the question of whether it is possible for a diagonalizing system, to learn to use environmental reward and punishment as an information, in order to appropriately adapt. More specifically, we study the possiblity of such a system, to learn to use diagonalization on the basis of a rewarding function. Relevant phenomena regarding memory are also investigated.

READ FULL TEXT
research
01/18/2022

Inducing Structure in Reward Learning by Learning Features

Reward learning enables robots to learn adaptable behaviors from human i...
research
06/19/2019

Adapting Behaviour via Intrinsic Reward: A Survey and Empirical Study

Learning about many things can provide numerous benefits to a reinforcem...
research
02/03/2021

Embodied Intelligence via Learning and Evolution

The intertwined processes of learning and evolution in complex environme...
research
07/12/2020

Learning Frame Level Attention for Environmental Sound Classification

Environmental sound classification (ESC) is a challenging problem due to...
research
07/05/2022

The StarCraft Multi-Agent Challenges+ : Learning of Multi-Stage Tasks and Environmental Factors without Precise Reward Functions

In this paper, we propose a novel benchmark called the StarCraft Multi-A...
research
11/06/2018

Blockchain and human episodic memory

We relate the concepts used in decentralized ledger technology to studie...
research
09/17/2021

Knowledge is reward: Learning optimal exploration by predictive reward cashing

There is a strong link between the general concept of intelligence and t...

Please sign up or login with your details

Forgot password? Click here to reset