Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization | Paper Citation Record

Observation cf65120c-5e39-42e5-b796-7692035685d2 · outbound

This paper cites Du, Wei Hu, Zhiyuan Li, and Ruosong Wang.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Du, Wei Hu, Zhiyuan Li, and Ruosong Wang

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:09.203822Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 83dd3c12-b08d-4f3a-81a2-57cb4c270077 · outbound

This paper cites an unresolved cited work.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Unresolved cited work

Reference 2

Resolution: unresolved; raw_fallback, observed 2026-08-07T13:13:08.939482Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 74c7d6ca-ef8d-41ef-a13f-d5ba1e3ad86e · outbound

This paper cites Penalising the biases in norm regularisation enforces sparsity https://papers.neurips.cc/paper_files/paper/2023/hash/b444ad72520a5f5c467343be88e352ed-Abstract-Conference.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Penalising the biases in norm regularisation enforces sparsity https://papers.neurips.cc/paper_files/paper/2023/hash/b444ad72520a5f5c467343be88e352ed-Abstract-Conference.html

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:08.668772Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 9c1ef4dc-9322-4aff-a239-1c5c4fb6f1ca · outbound

This paper cites Early alignment in two-layer networks training is a two-edged sword 10.48550/arxiv.2401.10791.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Early alignment in two-layer networks training is a two-edged sword 10.48550/arxiv.2401.10791

Reference 4

Resolution: verified exact; doi, observed 2026-08-07T13:13:01.118361Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation f6a9376d-c56c-4135-ae90-47898d9cfd9c · outbound

This paper cites Simplicity bias and optimization threshold in two-layer ReLU networks https://openreview.net/forum?id=qAarsvflTa.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Simplicity bias and optimization threshold in two-layer ReLU networks https://openreview.net/forum?id=qAarsvflTa

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:08.496137Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 792d70e4-35af-48d8-be78-090179834cc0 · outbound

This paper cites an unresolved cited work.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Unresolved cited work

Reference 6

Resolution: unresolved; raw_fallback, observed 2026-08-07T13:13:08.357491Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 8f0db553-e388-4387-8853-a948e9210d09 · outbound

This paper cites How Uniform Random Weights Induce Non-uniform Bias: Typical Interpolating Neural Networks Generalize with Narrow Teachers https://openreview.net/forum?id=3eHNvPHL9Z.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization How Uniform Random Weights Induce Non-uniform Bias: Typical Interpolating Neural Networks Generalize with Narrow Teachers https://openreview.net/forum?id=3eHNvPHL9Z

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:08.217059Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 9b1854d3-34d3-4136-9f96-688a9f2403c1 · outbound

This paper cites Convergence of gradient descent for deep neural networks 10.48550/arxiv.2203.16462.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Convergence of gradient descent for deep neural networks 10.48550/arxiv.2203.16462

Reference 8

Resolution: verified exact; doi, observed 2026-08-07T13:13:00.843877Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 03f525e9-9160-44a3-a045-08d516f2efe6 · outbound

This paper cites Loss Landscapes are All You Need: Neural Network Generalization Can Be Explained Without the Implicit Bias of Gradient Descent https://openreview.net/forum?id=QC10RmRbZy9.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Loss Landscapes are All You Need: Neural Network Generalization Can Be Explained Without the Implicit Bias of Gradient Descent https://openreview.net/forum?id=QC10RmRbZy9

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:08.001695Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 071cdf86-ea16-4c38-a119-33a3a0a780ff · outbound

This paper cites an unresolved cited work.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Unresolved cited work

Reference 10

Resolution: unresolved; raw_fallback, observed 2026-08-07T13:13:07.949186Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 38f3a440-21cb-4a86-8869-33e040ebb0ae · outbound

This paper cites an unresolved cited work.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Unresolved cited work

Reference 11

Resolution: unresolved; raw_fallback, observed 2026-08-07T13:13:07.918542Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 64cdbab3-38e7-43c0-a369-5bd2ede0cb6c · outbound

This paper cites an unresolved cited work.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Unresolved cited work

Reference 12

Resolution: unresolved; raw_fallback, observed 2026-08-07T13:13:07.701863Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation e30d8cd1-5d7c-4bdb-91ac-65168652e6dd · outbound

This paper cites an unresolved cited work.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Unresolved cited work

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:55.590174Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b4bd2716-0494-4d02-bfb8-c1da6de70724 · outbound

This paper cites Bach, and Loucas Pillaud - Vivien.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Bach, and Loucas Pillaud - Vivien

Reference 14

Resolution: verified exact; doi, observed 2026-08-07T13:13:00.630343Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation e8e52731-abf1-486f-b05d-36faeb95cf96 · outbound

This paper cites an unresolved cited work.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Unresolved cited work

Reference 15

Resolution: unresolved; raw_fallback, observed 2026-08-07T13:13:07.503025Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 79819969-13bd-4fc5-b1f9-a5e6b3778476 · outbound

This paper cites Kakade, and Jason D.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Kakade, and Jason D

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:55.969941Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 174a3a10-6587-46e7-8fa9-61538a6fc583 · outbound

This paper cites https://www.jmlr.org/papers/v17/15-408.html CVXPY : A P ython-embedded modeling language for convex optimization.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization https://www.jmlr.org/papers/v17/15-408.html CVXPY : A P ython-embedded modeling language for convex optimization

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:07.256033Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 28dc0c1e-59b1-41da-9bd7-e02db3e4f05f · outbound

This paper cites Hamprecht.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Hamprecht

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:06.951243Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 2e16810a-a1fd-4a45-a54e-25e522cbd06a · outbound

This paper cites Du, Xiyu Zhai, Barnab \' a s P \' o czos, and Aarti Singh.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Du, Xiyu Zhai, Barnab \' a s P \' o czos, and Aarti Singh

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:06.708498Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 0b01b0f9-19ea-4923-afb6-ec89ca45688d · outbound

This paper cites Convex Geometry and Duality of Over-parameterized Neural Networks http://jmlr.org/papers/v22/20-1447.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Convex Geometry and Duality of Over-parameterized Neural Networks http://jmlr.org/papers/v22/20-1447.html

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:06.540935Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation b7873b5f-3ba9-4ee0-8a70-c6cd7daa7e7c · outbound

This paper cites An introduction to probability theory and its applications, volume 2.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization An introduction to probability theory and its applications, volume 2

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:06.364081Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 51ada76a-9c7e-49b9-b168-d8626d6c5b38 · outbound

This paper cites Vetrov, and Andrew Gordon Wilson.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Vetrov, and Andrew Gordon Wilson

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:06.086766Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation f8d588e8-3ef4-433d-ae93-2cedfb5b5b9a · outbound

This paper cites https://openreview.net/forum?id=HgOJlxzB16 SGD Finds then Tunes Features in Two-Layer Neural Networks with near-Optimal Sample Complexity: A Case Study in the XOR problem.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization https://openreview.net/forum?id=HgOJlxzB16 SGD Finds then Tunes Features in Two-Layer Neural Networks with near-Optimal Sample Complexity: A Case Study in the XOR problem

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:05.911555Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 28a6c832-38c5-4355-9f20-5bb655161c75 · outbound

This paper cites Truth or backpropaganda? An empirical investigation of deep learning theory https://openreview.net/forum?id=HyxyIgHFvr.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Truth or backpropaganda? An empirical investigation of deep learning theory https://openreview.net/forum?id=HyxyIgHFvr

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:05.787591Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation febe0f18-62a5-4f05-a621-8c1522f71328 · outbound

This paper cites Clarabel: An interior-point solver for conic programs with quadratic objectives.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Clarabel: An interior-point solver for conic programs with quadratic objectives

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:56.938493Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a96253ce-1c46-4c1f-b9cc-8169e7197c6b · outbound

This paper cites Haeffele and Ren \' e Vidal.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Haeffele and Ren \' e Vidal

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:57.023287Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 12dd8075-579f-4650-b68b-bc03a8f58350 · outbound

This paper cites Piecewise linear activations substantially shape the loss surfaces of neural networks https://openreview.net/forum?id=B1x6BTEKwr.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Piecewise linear activations substantially shape the loss surfaces of neural networks https://openreview.net/forum?id=B1x6BTEKwr

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:05.676439Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation c6928d55-1f75-46db-bb7c-805e90b5bf41 · outbound

This paper cites Deep Residual Learning for Image Recognition 10.1109/cvpr.2016.90.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Deep Residual Learning for Image Recognition 10.1109/cvpr.2016.90

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:57.324053Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 752329bc-34da-4c9c-97ac-02aaa5a0259e · outbound

This paper cites Neural Tangent Kernel: Convergence and Generalization in Neural Networks https://proceedings.neurips.cc/paper/2018/hash/5a4be1fa34e62bb8a6ec6b91d2462f5a-Abstract.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Neural Tangent Kernel: Convergence and Generalization in Neural Networks https://proceedings.neurips.cc/paper/2018/hash/5a4be1fa34e62bb8a6ec6b91d2462f5a-Abstract.html

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:05.456856Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 16283734-4f5a-42f1-bc36-49195400c4fa · outbound

This paper cites an unresolved cited work.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Unresolved cited work

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:57.560687Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f4c597be-e459-448e-9c63-6d5b0870e528 · outbound

This paper cites Mildly Overparameterized ReLU Networks Have a Favorable Loss Landscape https://openreview.net/forum?id=10WARaIwFn.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Mildly Overparameterized ReLU Networks Have a Favorable Loss Landscape https://openreview.net/forum?id=10WARaIwFn

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:05.283568Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 46685940-4451-4b0e-8239-74ac682d9155 · outbound

This paper cites Deep Learning without Poor Local Minima https://proceedings.neurips.cc/paper/2016/hash/f2fc990265c712c49d51a18a32b39f0c-Abstract.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Deep Learning without Poor Local Minima https://proceedings.neurips.cc/paper/2016/hash/f2fc990265c712c49d51a18a32b39f0c-Abstract.html

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:05.082243Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation eee97ce2-6b4e-4dad-983b-60b3b5db89fb · outbound

This paper cites Exploring The Loss Landscape Of Regularized Neural Networks Via Convex Duality https://openreview.net/forum?id=4xWQS2z77v.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Exploring The Loss Landscape Of Regularized Neural Networks Via Convex Duality https://openreview.net/forum?id=4xWQS2z77v

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:04.832441Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 2dc68690-56d3-48db-8bf9-d4c0fb558842 · outbound

This paper cites Deep Linear Networks with Arbitrary Loss: All Local Minima Are Global http://proceedings.mlr.press/v80/laurent18a.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Deep Linear Networks with Arbitrary Loss: All Local Minima Are Global http://proceedings.mlr.press/v80/laurent18a.html

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:04.610229Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation e05247a5-00b8-498d-a4fc-57a01db631b3 · outbound

This paper cites Michaud, and Max Tegmark.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Michaud, and Max Tegmark

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:04.456563Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 79238af5-cf16-4216-898e-969faaaa50b1 · outbound

This paper cites Gradient Descent on Two-layer Nets: Margin Maximization and Simplicity Bias https://proceedings.neurips.cc/paper/2021/hash/6c351da15b5e8a743a21ee96a86e25df-Abstract.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Gradient Descent on Two-layer Nets: Margin Maximization and Simplicity Bias https://proceedings.neurips.cc/paper/2021/hash/6c351da15b5e8a743a21ee96a86e25df-Abstract.html

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:04.326855Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 7dcba60d-5130-4eb8-a06f-75bed81ef595 · outbound

This paper cites Lee, and Wei Hu.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Lee, and Wei Hu

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:04.151036Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 7f4f4e97-1a83-4c3e-95aa-f8f1cb223d96 · outbound

This paper cites Gradient Descent Quantizes ReLU Network Features.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Gradient Descent Quantizes ReLU Network Features

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:58.348208Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb55700f-f0c9-4f44-a56d-d5e518060c56 · outbound

This paper cites A mean field view of the landscape of two-layer neural networks 10.1073/pnas.1806579115.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization A mean field view of the landscape of two-layer neural networks 10.1073/pnas.1806579115

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:58.426905Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 349c5a1b-fa92-42ed-b34e-603a66389385 · outbound

This paper cites Early Neuron Alignment in Two-layer ReLU Networks with Small Initialization https://openreview.net/forum?id=QibPzdVrRu.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Early Neuron Alignment in Two-layer ReLU Networks with Small Initialization https://openreview.net/forum?id=QibPzdVrRu

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:03.976474Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 279e937b-4b17-45e4-bad8-8d78436d72b9 · outbound

This paper cites Optimal Sets and Solution Paths of ReLU Networks https://proceedings.mlr.press/v202/mishkin23a.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Optimal Sets and Solution Paths of ReLU Networks https://proceedings.mlr.press/v202/mishkin23a.html

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:03.780828Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 14ad19bc-d854-492b-9efd-5753fc7dc9e5 · outbound

This paper cites In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learning.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization In Search of the Real Inductive Bias: On the Role of Implicit Regularization in Deep Learning

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:58.738470Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b1bb096-e0cc-491c-b36e-c123b365e76b · outbound

This paper cites On Connected Sublevel Sets in Deep Learning http://proceedings.mlr.press/v97/nguyen19a.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization On Connected Sublevel Sets in Deep Learning http://proceedings.mlr.press/v97/nguyen19a.html

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:03.632268Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 9331022f-4f6d-48ce-b281-a344eee577ab · outbound

This paper cites A Note on Connectivity of Sublevel Sets in Deep Learning.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization A Note on Connectivity of Sublevel Sets in Deep Learning

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:58.964229Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d225fce-7953-4a55-ae08-b58b9f4cdc38 · outbound

This paper cites When Are Solutions Connected in Deep Networks? https://proceedings.neurips.cc/paper/2021/hash/af5baf594e9197b43c9f26f17b205e5b-Abstract.html In NeurIPS, pages 20956--20969, 2021.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization When Are Solutions Connected in Deep Networks? https://proceedings.neurips.cc/paper/2021/hash/af5baf594e9197b43c9f26f17b205e5b-Abstract.html In NeurIPS, pages 20956--20969, 2021

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:03.463658Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation ded24de2-dbb3-4229-b5a3-351500f6c86a · outbound

This paper cites Banach space representer theorems for neural networks and ridge splines https://dl.acm.org/doi/10.5555/3546258.3546301.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Banach space representer theorems for neural networks and ridge splines https://dl.acm.org/doi/10.5555/3546258.3546301

Reference 46

Resolution: metadata mismatch; raw_fallback, observed 2026-08-07T13:13:01.452769Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 027e40ea-cedb-4cb4-bbe5-9e2f036d8106 · outbound

This paper cites Neural Networks are Convex Regularizers: Exact Polynomial-time Convex Optimization Formulations for Two-layer Networks http://proceedings.mlr.press/v119/pilanci20a.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Neural Networks are Convex Regularizers: Exact Polynomial-time Convex Optimization Formulations for Two-layer Networks http://proceedings.mlr.press/v119/pilanci20a.html

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:03.250869Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation af995fc2-0e66-43c6-827f-2fdf0ae520e5 · outbound

This paper cites Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:59.316611Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 147fe6d0-4954-400d-b601-98d8cdd4522c · outbound

This paper cites Trainability and accuracy of artificial neural networks: An interacting particle system approach 10.1002/cpa.22074.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Trainability and accuracy of artificial neural networks: An interacting particle system approach 10.1002/cpa.22074

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:59.419770Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7b54f09-d21b-4f87-9ddd-80e61a9a7a81 · outbound

This paper cites Spurious Local Minima are Common in Two-Layer ReLU Neural Networks http://proceedings.mlr.press/v80/safran18a.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Spurious Local Minima are Common in Two-Layer ReLU Neural Networks http://proceedings.mlr.press/v80/safran18a.html

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:03.111915Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation f60e7662-4104-42ba-b431-718fd99987c1 · outbound

This paper cites How do infinite width bounded norm networks look in function space? http://proceedings.mlr.press/v99/savarese19a.html In COLT, pages 2667--2690, 2019.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization How do infinite width bounded norm networks look in function space? http://proceedings.mlr.press/v99/savarese19a.html In COLT, pages 2667--2690, 2019

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:02.939431Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 52ee293f-9430-491e-915b-45413cc09286 · outbound

This paper cites Understanding machine learning: From theory to algorithms 10.1017/CBO9781107298019.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Understanding machine learning: From theory to algorithms 10.1017/CBO9781107298019

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:59.653892Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3a176d66-6b8b-45b8-83d3-617a6339a7b5 · outbound

This paper cites Jamaloddin Golestani.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Jamaloddin Golestani

Reference 53

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:02.705425Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 0855d4a2-ac27-4f9b-bfdc-55c07735c983 · outbound

This paper cites Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances http://proceedings.mlr.press/v139/simsek21a.html.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances http://proceedings.mlr.press/v139/simsek21a.html

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:02.467783Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 8fc4d611-2594-4d34-ab6c-1d366d687115 · outbound

This paper cites The Global Landscape of Neural Networks: An Overview 10.1109/msp.2020.3004124.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization The Global Landscape of Neural Networks: An Overview 10.1109/msp.2020.3004124

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-07T13:12:59.866232Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 42386525-5e0c-41b5-b985-efdf1ddaf1e1 · outbound

This paper cites Bandeira, and Joan Bruna.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Bandeira, and Joan Bruna

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:02.276608Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation c461e396-90eb-48de-954c-3a24e8ad053d · outbound

This paper cites The Hidden Convex Optimization Landscape of Regularized Two-Layer ReLU Networks: an Exact Characterization of Optimal Solutions https://openreview.net/forum?id=Z7Lk2cQEG8a.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization The Hidden Convex Optimization Landscape of Regularized Two-Layer ReLU Networks: an Exact Characterization of Optimal Solutions https://openreview.net/forum?id=Z7Lk2cQEG8a

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:02.091759Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 064d0a4b-9a0b-4ce9-a5ac-6bb0e04fee23 · outbound

This paper cites On the Convergence of Gradient Descent Training for Two-layer ReLU-networks in the Mean Field Regime.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization On the Convergence of Gradient Descent Training for Two-layer ReLU-networks in the Mean Field Regime

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-07T13:13:00.129590Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 047842c8-cae0-4b6f-88e8-63aebf6436ef · outbound

This paper cites Woodworth, Suriya Gunasekar, Jason D.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Woodworth, Suriya Gunasekar, Jason D

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:01.882611Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation a1e10686-1355-4d63-8293-7c13b534aa61 · outbound

This paper cites Small nonlinearities in activation functions create bad local minima in neural networks https://openreview.net/forum?id=rke\_YiRct7.

Favorability of Loss Landscape with Weight Decay Requires Both Large Overparametrization and Initialization Small nonlinearities in activation functions create bad local minima in neural networks https://openreview.net/forum?id=rke\_YiRct7

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T13:13:01.699375Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Coverage vector

Reference resolution

Outbound references

Pith citing papers