sulcantonin/torchmodal - 5 Help Wanted Issues

Open Issues Need Help

Add a worked-example page to the documentation site 6 days ago

documentation good first issue

Differentiable modal logic for PyTorch: □ and ♢ as trainable layers over a Kripke structure, with sound interval bounds and a learnable accessibility relation. Oral @ NeSy 2026.

Python

#differentiable-logic#epistemic-logic#fuzzy-logic#kripke-semantics#logic#machine-learning#modal-logic#neural-symbolic#neurosymbolic#neurosymbolic-ai#pytorch#temporal-logic

Seed torch in the example scripts that only seed numpy 6 days ago

bug good first issue

sulcantonin/torchmodal

Differentiable modal logic for PyTorch: □ and ♢ as trainable layers over a Kripke structure, with sound interval bounds and a learnable accessibility relation. Oral @ NeSy 2026.

Python

#differentiable-logic#epistemic-logic#fuzzy-logic#kripke-semantics#logic#machine-learning#modal-logic#neural-symbolic#neurosymbolic#neurosymbolic-ai#pytorch#temporal-logic

Add a seriality regulariser to losses.AxiomRegularization 6 days ago

enhancement good first issue

sulcantonin/torchmodal

Differentiable modal logic for PyTorch: □ and ♢ as trainable layers over a Kripke structure, with sound interval bounds and a learnable accessibility relation. Oral @ NeSy 2026.

Python

#differentiable-logic#epistemic-logic#fuzzy-logic#kripke-semantics#logic#machine-learning#modal-logic#neural-symbolic#neurosymbolic#neurosymbolic-ai#pytorch#temporal-logic

Add an add_until_graph node to inference.FormulaGraph 6 days ago

enhancement good first issue

sulcantonin/torchmodal

Differentiable modal logic for PyTorch: □ and ♢ as trainable layers over a Kripke structure, with sound interval bounds and a learnable accessibility relation. Oral @ NeSy 2026.

Python

#differentiable-logic#epistemic-logic#fuzzy-logic#kripke-semantics#logic#machine-learning#modal-logic#neural-symbolic#neurosymbolic#neurosymbolic-ai#pytorch#temporal-logic

Add an nn.UntilGraph module wrapper for functional.until_graph 6 days ago

enhancement good first issue

sulcantonin/torchmodal

Differentiable modal logic for PyTorch: □ and ♢ as trainable layers over a Kripke structure, with sound interval bounds and a learnable accessibility relation. Oral @ NeSy 2026.

Python

#differentiable-logic#epistemic-logic#fuzzy-logic#kripke-semantics#logic#machine-learning#modal-logic#neural-symbolic#neurosymbolic#neurosymbolic-ai#pytorch#temporal-logic