mlsanityData to 7 Oct 2026

IID Relaxation by Logical Expressivity: A Research Agenda for Fitting Logics to Neurosymbolic Requirements

arXiv 2404.19485 ↗First posted 30 Apr 2024cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Neurosymbolic background knowledge and the expressivity required of its logic can break Machine Learning assumptions about data Independence and Identical Distribution. In this position paper we propose to analyze IID relaxation in a hierarchy of logics that fit different use case requirements. We discuss the benefits of exploiting known data dependencies and distribution constraints for Neurosymbolic use cases and argue that the expressivity required for this knowledge has implications for the design of underlying ML routines. This opens a new research agenda with general questions about Neurosymbolic background knowledge and the expressivity required of its logic.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 23 of its 51 references that are in the map. First posted on arXiv 30 Apr 2024.

Builds on

23 of 51 references.

References · 23

  1. 2017Modeling Relational Data with Graph Convolutional Networkscited by 1,684
  2. 2016CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoningcited by 1,445
  3. 2016Optimization Methods for Large-Scale Machine Learningcited by 1,140
  4. 2019Neural Probabilistic Logic Programming in DeepProbLogcited by 389
  5. 2021Federated Learning on Non-IID Data: A Surveycited by 345
  6. 2017A Semantic Loss Function for Deep Learning with Symbolic Knowledgecited by 267
  7. 2020Neurosymbolic AI: The 3rd Wavecited by 253
  8. 2018Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shiftcited by 240
  9. 2021Towards Causal Representation Learningcited by 207
  10. 2017Neural-Symbolic Learning and Reasoning: A Survey and Interpretationcited by 195
  11. 2020The Next Decade in AI: Four Steps Towards Robust Artificial Intelligencecited by 194
  12. 2020The Computational Limits of Deep Learningcited by 188
11 more
  1. 2021The Benchmark Lotterycited by 133
  2. 2022Federated Learning on Non-IID Graphs via Structural Knowledge Sharingcited by 88
  3. 2021From Statistical Relational to Neurosymbolic Artificial Intelligence: a Surveycited by 72
  4. 2022Client Selection in Federated Learning: Principles, Challenges, and Opportunitiescited by 62
  5. 2020Closed Loop Neural-Symbolic Learning via Integrating Neural Perception, Grammar Parsing, and Symbolic Reasoningcited by 58
  6. 2021Modular Design Patterns for Hybrid Learning and Reasoning Systems: a taxonomy, patterns and use casescited by 39
  7. 2023A survey and taxonomy of loss functions in machine learningcited by 15
  8. 2013Querying the Guarded Fragmentcited by 6
  9. 2021Partially Observed Exchangeable Modelingcited by 3
  10. 2020Exploiting non-i.i.d. data towards more robust machine learning algorithmscited by 1
  11. 2022Modular design patterns for neural-symbolic integration: refinement and combinationcited by 1

Built on by

Cited by year

Citing papers per year: 2024: 1.

1 paper.

  1. 2024ULLER: A Unified Language for Learning and Reasoningcited by 9