We lifted 1 functions out of this paper's own repositories and ran 1 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 |
|---|---|---|
| alexrame/fishr | canonical | 1 of 1 |
| Function | Status | Where it lives |
|---|---|---|
| format_mean | Ran | alexrame/fishr/domainbed/scripts/collect_results.py pointer only (licence: GPL-3.0) · get_code("d6d6cc53be32629d") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Learning robust models that generalize well under changes in the data distribution is critical for real-world applications. To this end, there has been a growing surge of interest to learn simultaneously from multiple training domains - while enforcing different types of invariance across those domains. Yet, all existing approaches fail to show systematic benefits under controlled evaluation protocols. In this paper, we introduce a new regularization - named Fishr - that enforces domain invariance in the space of the gradients of the loss: specifically, the domain-level variances of gradients are matched across training domains. Our approach is based on the close relations between the gradient covariance, the Fisher Information and the Hessian of the loss: in particular, we show that Fishr eventually aligns the domain-level loss landscapes locally around the final weights. Extensive experiments demonstrate the effectiveness of Fishr for out-of-distribution generalization. Notably, Fishr improves the state of the art on the DomainBed benchmark and performs consistently better than Empirical Risk Minimization. Our code is available at https://github.com/alexrame/fishr.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2109.02934")
get_code_for_paper("2109.02934")
have("2109.02934")
Connect an agent — have() is free.