v159: Proceedings of IWSSL 2021

2026-06-10 · Source: Proceedings of Machine Learning Research · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems · Depth: Expert, quick

Summary

Volume 159 presents the proceedings of the Second International Workshop on Self-Supervised Learning (IWSSL), held virtually on August 13-14, 2021, and edited by Kristinn R. Thórisson and Paul Robertson. The workshop features diverse research, including Thórisson's "Explanation Hypothesis" in general self-supervised learning, and Pei Wang's proposal for a unified model of reasoning and learning. Contributions also cover empirical studies, such as Leonard M. Eberding's comparison of machine learners on an ABA experiment format of the Cart-Pole Task. Furthermore, Paul Robertson explores the role of "Artificial Emotions" for rapid online explorative learning, showcasing the breadth of innovative approaches within the self-supervised learning domain.

Key takeaway

The Second International Workshop on Self-Supervised Learning (IWSSL) Volume 159 presents diverse research advancing SSL through theoretical foundations and novel mechanisms. Key contributions include the "Explanation Hypothesis," a unified model of reasoning and learning, empirical comparisons on tasks like Cart-Pole, and the application of artificial emotions. This collection offers essential insights for AI/ML researchers and practitioners focused on developing more robust and autonomous learning systems.

Topics

• Self-Supervised Learning
• Explanation Hypothesis
• Reasoning and Learning Models
• Machine Learner Comparison
• Cart-Pole Task

Code references

• mlresearch/v159

Best for: AI Scientist, AI Student, Machine Learning Engineer

Related on AIssential

• SSL-R1: Self-Supervised Visual Reinforcement Post-Training for Multimodal Large Language Models — SSL-R1 uses self-supervised visual puzzles to provide scalable, verifiable rewards for MLLM reinforcement learning.
• Explaining Neural Networks in Preference Learning: a Post-hoc Inductive Logic Programming Approach — ILASP can approximate black-box Neural Networks for preference learning, even in high-dimensional spaces.
• v162: Proceedings of ICML 2022 — Volume 162 of the International Conference on Machine Learning (ICML), held in Baltimore, Maryland, USA, from July 17-23, 2022, presents a comprehensive collection of research across diverse machine learning domains.
• On the Expressive Power and Limitations of Multi-Layer SSMs — Multi-layer SSMs have inherent compositional limitations, but online Chain-of-Thought reasoning can make them as powerful as streaming algorithms.
• Peer-Predictive Self-Training for Language Model Reasoning — Aggregating peer predictions and weighting self-training by mutual information enables unsupervised language model improvement.
• PAINT: Partial-Solution Adaptive Interpolated Training for Self-Distilled Reasoners — PAINT improves LLM reasoning by adaptively masking solutions and interpolating energy at entropy-mismatch tokens.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Proceedings of Machine Learning Research.