Learning Loop

Learning loops encompass iterative processes where a model improves its performance by repeatedly learning from its own predictions or actions. Current research focuses on enhancing the efficiency and effectiveness of these loops, particularly within the context of large language models and meta-learning algorithms, exploring techniques like gradient sharing and self-supervised learning to accelerate training and mitigate issues like catastrophic forgetting. This research is significant because it addresses fundamental limitations in current AI systems, paving the way for more adaptable, efficient, and self-improving models with applications across diverse fields.

Papers

February 14, 2024

Into the Unknown: Self-Learning Large Language Models
Teddy Ferdinan, Jan Kocoń, Przemysław Kazienko
LLM Based Self Directed Learning Learning Loop

December 13, 2023

Accelerating Meta-Learning by Sharing Gradients
Oscar Chang, Hod Lipson
Multi Task Learning Meta Learning Gradient Based Meta Learning Gradient Sharing Learning Loop Encode Task

October 20, 2023

Learning to (Learn at Test Time)
Yu Sun, Xinhao Li, Karan Dalal, Chloe Hsu, Sanmi Koyejo, Carlos Guestrin, Xiaolong Wang, Tatsunori Hashimoto, Xinlei Chen
LeArning Abstract Self Attention Layer Test Time Linear Attention Self Supervised Task Learning Loop

September 15, 2023

Chain-of-Thought Reasoning is a Policy Improvement Operator
Hugh Zhang, David C. Parkes
Large Language Model Language Model Training Data Complex Reasoning Chain of Thought Learning Loop

July 22, 2023

Emergence of Adaptive Circadian Rhythms in Deep Reinforcement Learning
Aqeel Labash, Florian Fletzer, Daniel Majoral, Raul Vicente
Deep Reinforcement Learning Adaptive Importance Path Breaking Emergence Artificial Neuron Neuronal Dynamic Learning Loop

May 22, 2023

Mitigating Catastrophic Forgetting for Few-Shot Spoken Word Classification Through Meta-Learning
Ruan van der Merwe, Herman Kamper
Natural Language Catastrophic Forgetting Meta Learning Shot Classification Continual Learning Scenario Semantic Class Speech Classification Learning Loop

March 14, 2023

Understanding the Uncertainty Loop of Human-Robot Interaction
Jan Leusmann, Chao Wang, Michael Gienger, Albrecht Schmidt, Sven Mayer
Human Robot Interaction Assistive Robot Human Robot Robot Assistance Learning Loop Human Human

February 15, 2023

Streamlining models with explanations in the learning loop
Francesco Lomuscio, Paolo Bajardi, Alan Perotti, Elvio G. Amparore
Line by Line Explanation Black Box Black Box Model Local Explanation Post Hoc Explanation Learning Loop Explainable AI Approach

January 27, 2023

Alien Coding
Thibault Gauthier, Miroslav Olšák, Josef Urban
Neural Machine Translation Functional Programming Learning Loop

December 28, 2022

Wormhole MAML: Meta-Learning in Glued Parameter Space
Chih-Jung Tracy Chang, Yuan Gao, Beicheng Lou
Object Classification Parameter Space MNIST Canadian Institute for Advanced Expressivity Transfer Learning Loop

March 18, 2022

Negative Inner-Loop Learning Rates Learn Universal Features
Tom Starshak
Learning Rate Task Adaptation Meta Gradient Agnostic Meta learnIng Negative Learning Learning Loop