Replication Data for: On the Universality of the Double Descent Peak in Ridgeless Regression (doi:10.18419/darus-1771)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

Replication Data for: On the Universality of the Double Descent Peak in Ridgeless Regression

doi:10.18419/darus-1771

DaRUS

2021-03-23

1

Holzmüller, David, 2021, "Replication Data for: On the Universality of the Double Descent Peak in Ridgeless Regression", https://doi.org/10.18419/darus-1771, DaRUS, V1

Replication Data for: On the Universality of the Double Descent Peak in Ridgeless Regression

doi:10.18419/darus-1771

Holzmüller, David (Universität Stuttgart)

EXC 2075 - 390740016

DaRUS

Holzmüller, David

Holzmüller, David

Steinwart, Ingo

Holzmüller, David

2021-03-16

https://doi.org/10.18419/darus-1771

Computer and Information Science, Mathematical Sciences, Computer and Information Science, Mathematical Sciences

This dataset contains code used to generate the figures in the paper <em>On the Universality of the Double Descent Peak in Ridgeless Regression, David Holzmüller, International Conference on Learning Representations 2021</em>. The code is also provided on <a href=https://github.com/dholzmueller/universal_double_descent>GitHub</a>. Here, we additionally provide the data that is generated by the code and that is required to generate the plots. To use this data, simply unpack the (large!) file <code>data.tar.gz</code> inside the folder where the Python files are located. If the data is not downloaded and unpacked, the code will automatically compute it, which can take about one day on a 6-core-CPU. Information on the code and used software can be found in the file README.md.

Licensed under the <a href="http://www.apache.org/licenses/LICENSE-2.0">Apache License, Version 2.0</a> (the "License"); Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See <a href="http://www.apache.org/licenses/LICENSE-2.0">the License</a> for the specific language governing permissions and limitations under the License.

2010.01851

David Holzmüller. On the Universality of the Double Descent Peak in Ridgeless Regression. International Conference on Learning Representations, 2021.

Label:

computation.py

text/x-python

Label:

data.tar.gz

Text:

Data generated by running the code

application/gzip

Label:

main.py

text/x-python

Label:

plotting.py

text/x-python

Label:

README.md

text/markdown

Label:

sampling.py

text/x-python

Label:

utils.py

text/x-python