We lifted 5 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| pnnl/modelsym | canonical | 2 of 5 |
| Function | Status | Where it lives |
|---|---|---|
| sinkhorn | Ran | pnnl/modelsym/model_symmetries/stitching/stitching.py pointer only (licence: NOASSERTION) · get_code("57f3995b8ea26d10") |
| softsort | Ran | pnnl/modelsym/model_symmetries/stitching/stitching.py pointer only (licence: NOASSERTION) · get_code("fe2339f3b4161fc7") |
| get_stitch_model | Not yet run | pnnl/modelsym/model_symmetries/stitching/stitching.py pointer only (licence: NOASSERTION) · get_code("4592fbb994b52a15") |
| wreath_cka | Not yet run | pnnl/modelsym/model_symmetries/alignment/alignment.py pointer only (licence: NOASSERTION) · get_code("cfa2e7f47dbee374") |
| wreath_procrustes | Not yet run | pnnl/modelsym/model_symmetries/alignment/alignment.py pointer only (licence: NOASSERTION) · get_code("7302057fbb1a0c89") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Symmetry is a fundamental tool in the exploration of a broad range of complex systems. In machine learning symmetry has been explored in both models and data. In this paper we seek to connect the symmetries arising from the architecture of a family of models with the symmetries of that family's internal representation of data. We do this by calculating a set of fundamental symmetry groups, which we call the intertwiner groups of the model. We connect intertwiner groups to a model's internal representations of data through a range of experiments that probe similarities between hidden states across models with the same architecture. Our work suggests that the symmetries of a network are propagated into the symmetries in that network's representation of data, providing us with a better understanding of how architecture affects the learning and prediction process. Finally, we speculate that for ReLU networks, the intertwiner groups may provide a justification for the common practice of concentrating model interpretability exploration on the activation basis in hidden layers rather than arbitrary linear combinations thereof.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2205.14258")
get_code_for_paper("2205.14258")
have("2205.14258")
Connect an agent — have() is free.