Exploration Degree Bias: The Hidden Influence of Node Degree in Graph Neural Network-Based Reinforcement Learning
Graph Neural Networks (GNNs) have demonstrated remarkable performance in tasks involving graph-structured data, but they also exhibit biases linked to node degrees. This paper explores a specific manifestation of such bias, termed Exploration Degree Bias (EDB), in the context of Reinforcement Learni...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10838508/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832592928830128128 |
|---|---|
| author | Peter Tarabek David Matis |
| author_facet | Peter Tarabek David Matis |
| author_sort | Peter Tarabek |
| collection | DOAJ |
| description | Graph Neural Networks (GNNs) have demonstrated remarkable performance in tasks involving graph-structured data, but they also exhibit biases linked to node degrees. This paper explores a specific manifestation of such bias, termed Exploration Degree Bias (EDB), in the context of Reinforcement Learning (RL). We show that EDB arises from the inherent design of GNNs, where nodes with high or low degrees disproportionately influence output logits used for decision-making. This phenomenon impacts exploration in RL, skewing it away from mid-degree nodes, potentially hindering the discovery of optimal policies. We provide a systematic investigation of EDB across widely used GNN architectures—GCN, GraphSAGE, GAT, and GIN—by quantifying correlations between node degrees and logits. Our findings reveal that EDB varies by architecture and graph configuration, with GCN and GIN exhibiting the strongest biases. Moreover, analysis of DQN and PPO RL agents illustrates how EDB can distort exploration patterns, with DQN exhibiting EDB under low exploration rates and PPO showing a partial ability to counteract these effects through its probabilistic sampling mechanism. Our contributions include defining and quantifying EDB, providing experimental insights into its existence and variability, and analyzing its implications for RL. These findings underscore the need to address degree-related biases in GNNs to enhance RL performance on graph-based tasks. |
| format | Article |
| id | doaj-art-66dd5a03c2904cb68e3b90679df6ebf8 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-66dd5a03c2904cb68e3b90679df6ebf82025-01-21T00:01:03ZengIEEEIEEE Access2169-35362025-01-0113107461075710.1109/ACCESS.2025.352887810838508Exploration Degree Bias: The Hidden Influence of Node Degree in Graph Neural Network-Based Reinforcement LearningPeter Tarabek0https://orcid.org/0000-0002-4815-3933David Matis1https://orcid.org/0009-0007-9023-5725Faculty of Management Science and Informatics, University of Žilina, Žilina, SlovakiaFaculty of Management Science and Informatics, University of Žilina, Žilina, SlovakiaGraph Neural Networks (GNNs) have demonstrated remarkable performance in tasks involving graph-structured data, but they also exhibit biases linked to node degrees. This paper explores a specific manifestation of such bias, termed Exploration Degree Bias (EDB), in the context of Reinforcement Learning (RL). We show that EDB arises from the inherent design of GNNs, where nodes with high or low degrees disproportionately influence output logits used for decision-making. This phenomenon impacts exploration in RL, skewing it away from mid-degree nodes, potentially hindering the discovery of optimal policies. We provide a systematic investigation of EDB across widely used GNN architectures—GCN, GraphSAGE, GAT, and GIN—by quantifying correlations between node degrees and logits. Our findings reveal that EDB varies by architecture and graph configuration, with GCN and GIN exhibiting the strongest biases. Moreover, analysis of DQN and PPO RL agents illustrates how EDB can distort exploration patterns, with DQN exhibiting EDB under low exploration rates and PPO showing a partial ability to counteract these effects through its probabilistic sampling mechanism. Our contributions include defining and quantifying EDB, providing experimental insights into its existence and variability, and analyzing its implications for RL. These findings underscore the need to address degree-related biases in GNNs to enhance RL performance on graph-based tasks.https://ieeexplore.ieee.org/document/10838508/Exploration degree biasgraph neural networksmessage-passing mechanismnode degree biasreinforcement learning |
| spellingShingle | Peter Tarabek David Matis Exploration Degree Bias: The Hidden Influence of Node Degree in Graph Neural Network-Based Reinforcement Learning IEEE Access Exploration degree bias graph neural networks message-passing mechanism node degree bias reinforcement learning |
| title | Exploration Degree Bias: The Hidden Influence of Node Degree in Graph Neural Network-Based Reinforcement Learning |
| title_full | Exploration Degree Bias: The Hidden Influence of Node Degree in Graph Neural Network-Based Reinforcement Learning |
| title_fullStr | Exploration Degree Bias: The Hidden Influence of Node Degree in Graph Neural Network-Based Reinforcement Learning |
| title_full_unstemmed | Exploration Degree Bias: The Hidden Influence of Node Degree in Graph Neural Network-Based Reinforcement Learning |
| title_short | Exploration Degree Bias: The Hidden Influence of Node Degree in Graph Neural Network-Based Reinforcement Learning |
| title_sort | exploration degree bias the hidden influence of node degree in graph neural network based reinforcement learning |
| topic | Exploration degree bias graph neural networks message-passing mechanism node degree bias reinforcement learning |
| url | https://ieeexplore.ieee.org/document/10838508/ |
| work_keys_str_mv | AT petertarabek explorationdegreebiasthehiddeninfluenceofnodedegreeingraphneuralnetworkbasedreinforcementlearning AT davidmatis explorationdegreebiasthehiddeninfluenceofnodedegreeingraphneuralnetworkbasedreinforcementlearning |