We lifted 4 functions out of this paper's own repositories and ran 4 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 |
|---|---|---|
| kisnikser/landscape-hessian | canonical | 4 of 4 |
| Function | Status | Where it lives |
|---|---|---|
| calculate_ema | Ran | kisnikser/landscape-hessian/code/get_loss_values.py code served (permissive licence) · get_code("cf5d50a254b9b86e") |
| calculate_ema | Ran | kisnikser/landscape-hessian/code/plot_differences.py code served (permissive licence) · get_code("a7406e6dd4491589") |
| criterion_params | Ran | kisnikser/landscape-hessian/code/get_loss_values.py code served (permissive licence) · get_code("549ca336ff33b04d") |
| get_loss_abs_differences | Ran | kisnikser/landscape-hessian/code/get_loss_values.py code served (permissive licence) · get_code("1c3c27dd0b929d96") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
The loss landscape of neural networks is a critical aspect of their training, and understanding its properties is essential for improving their performance. In this paper, we investigate how the loss surface changes when the sample size increases, a previously unexplored issue. We theoretically analyze the convergence of the loss landscape in a fully connected neural network and derive upper bounds for the difference in loss function values when adding a new object to the sample. Our empirical study confirms these results on various datasets, demonstrating the convergence of the loss function surface for image classification tasks. Our findings provide insights into the local geometry of neural loss landscapes and have implications for the development of sample size determination techniques.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2409.11995")
get_code_for_paper("2409.11995")
have("2409.11995")
Connect an agent — have() is free.