Initial upload: 200 pretrain + 2000 oracle models, logits, margins, and 10 forget sets

LICENSE

@@ -0,0 +1,22 @@
+MIT License
+
+Copyright (c) 2025 Pol Puigdemont, Roy Rinberg, Martin Pawelczyk, Volkan Cevher
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+

README.md

@@ -0,0 +1,174 @@
+---
+license: mit
+tags:
+  - machine-unlearning
+  - cifar10
+  - computer-vision
+  - robustness
+  - benchmarks
+---
+
+## EasyDUB Dataset
+
+Precomputed CIFAR-10 data for KLOM (KL-divergence of Margins) evaluation of data-unlearning methods.
+
+This dataset contains:
+- **200 pretrain models**: ResNet9 models trained on the full CIFAR-10 training set (50,000 samples).
+- **200 oracle models per forget set**: ResNet9 models retrained on the retain set (train minus forget) for each of 10 forget sets.
+- **Logits and margins**: Precomputed logits and margins for all models on train/val/forget/retain splits.
+
+All models are checkpointed at epoch 23 (out of 24 total training epochs).
+
+### Directory structure
+
+The on-disk layout is:
+
+```text
+EasyDUB-dataset/
+├── models/
+│   └── cifar10/
+│       ├── pretrain/
+│       │   └── resnet9/
+│       │       └── id_X_epoch_23.pt          # 200 models (X = 0–199)
+│       └── oracle/
+│           └── forget_Z/
+│               └── resnet9/
+│                   └── id_X_epoch_23.pt      # 200 models per forget set
+│
+├── logits/
+│   └── cifar10/
+│       ├── pretrain/
+│       │   ├── retain/
+│       │   │   └── resnet9/
+│       │   │       └── id_X_epoch_23.npy     # Full train set logits
+│       │   ├── val/
+│       │   │   └── resnet9/
+│       │   │       └── id_X_epoch_23.npy     # Validation logits
+│       │   └── forget_Z/
+│       │       └── resnet9/
+│       │           └── id_X_epoch_23.npy     # Forget-set logits
+│       └── oracle/
+│           └── forget_Z/
+│               ├── retain/
+│               │   └── resnet9/
+│               │       └── id_X_epoch_23.npy # Retain logits
+│               ├── forget/
+│               │   └── resnet9/
+│               │       └── id_X_epoch_23.npy # Forget logits
+│               └── val/
+│                   └── resnet9/
+│                       └── id_X_epoch_23.npy # Validation logits
+│
+├── margins/
+│   └── cifar10/
+│       └── [same structure as logits/]
+│
+└── forget_sets/
+    └── cifar10/
+        └── forget_set_Z.npy                  # Indices into CIFAR-10 train set
+```
+
+### File naming
+
+- **Models**: `id_{MODEL_ID}_epoch_{EPOCH}.pt` (e.g. `id_42_epoch_23.pt`)
+- **Logits / margins**: `id_{MODEL_ID}_epoch_{EPOCH}.npy`
+- **Forget sets**: `forget_set_{SET_ID}.npy` (e.g. `forget_set_1.npy`)
+
+### Shapes and dtypes
+
+- **Logits**: `(n_samples, 10)` NumPy arrays of `float32` — raw model outputs for the 10 CIFAR-10 classes.
+- **Margins**: `(n_samples,)` NumPy arrays of `float32` — scalar margins (see formula below).
+- **Forget sets**: `(n_forget_samples,)` NumPy arrays of integer indices into the CIFAR-10 training set in `[0, 49_999]`.
+
+Typical sizes:
+- Train set: 50_000 samples
+- Validation set: 10_000 samples
+- Forget sets: 10–1000 samples (varies by set)
+
+### Margin definition
+
+For each sample with logits `logits` and true label `true_label`:
+
+```python
+import torch
+
+def compute_margin(logits: torch.Tensor, true_label: int) -> torch.Tensor:
+    logit_other = logits.clone()
+    logit_other[true_label] = -torch.inf
+    return logits[true_label] - logit_other.logsumexp(dim=-1)
+```
+
+Higher margins indicate higher confidence in the correct class relative to all others (via log-sum-exp).
+
+### Forget sets
+
+The dataset includes 10 CIFAR-10 forget sets:
+
+- **Forget set 1**: 10 random samples
+- **Forget set 2**: 100 random samples
+- **Forget set 3**: 1_000 random samples
+- **Forget set 4**: 10 samples with highest projection onto the 1st principal component
+- **Forget set 5**: 100 samples with highest projection onto the 1st principal component
+- **Forget set 6**: 250 samples with highest + 250 with lowest projection onto the 1st principal component
+- **Forget set 7**: 10 samples with highest projection onto the 2nd principal component
+- **Forget set 8**: 100 samples with highest projection onto the 2nd principal component
+- **Forget set 9**: 250 samples with highest + 250 with lowest projection onto the 2nd principal component
+- **Forget set 10**: 100 samples closest in CLIP image space to a reference cassowary image
+
+Each `forget_set_Z.npy` is a 1D array of training indices.
+
+### Quick start
+
+The companion [EasyDUB-code](https://github.com/easydub/EasyDUB-code) repository provides utilities and unlearning methods on top of this dataset.
+
+Here is a minimal example using only NumPy and PyTorch:
+
+```python
+import numpy as np
+import torch
+
+root = "EasyDUB-dataset"
+
+# Load margins for a single pretrain model on the validation set
+margins = np.load(f"{root}/margins/cifar10/pretrain/val/resnet9/id_0_epoch_23.npy")
+
+# Load oracle margins for the same model index and forget set (example: forget_set_1)
+oracle_margins = np.load(
+    f"{root}/margins/cifar10/oracle/forget_1/val/resnet9/id_0_epoch_23.npy"
+)
+
+print(margins.shape, oracle_margins.shape)
+```
+
+For a higher-level end-to-end demo (including unlearning methods and KLOM computation), see the [EasyDUB-code](https://github.com/easydub/EasyDUB-code) GitHub repository. In particular, `strong_test.py` in `EasyDUB-code` runs a reproducible noisy-SGD unlearning experiment comparing:
+
+- `KLOM(pretrain, oracle)`
+- `KLOM(noisy_descent, oracle)`
+
+### Training procedure (summary)
+
+All pretrain and oracle models share the same training setup:
+
+- Optimizer: SGD with momentum
+- Learning rate: 0.4 (triangular schedule peaking at epoch 5)
+- Momentum: 0.9
+- Weight decay: 5e-4
+- Epochs: 24 total, checkpoint used here is epoch 23
+- Mixed precision: enabled (FP16)
+- Label smoothing: 0.0
+
+Pretrain models are trained on the full CIFAR-10 training set. Oracle models are trained on the retain set (training set minus the corresponding forget set) for each forget set.
+
+### Citation
+
+If you use EasyDUB in your work, please cite:
+
+```bibtex
+@inproceedings{puigdemont2025dataunlearnbench,
+  title     = {Data-Unlearn-Bench: Making Evaluating Data Unlearning Easy},
+  author    = {Puigdemont, Pol and Rinberg, Roy and Pawelczyk, Martin and Cevher, Volkan},
+  booktitle = {MUGEN Workshop at ICML},
+  year      = {2025},
+}
+```
+

forget_sets/cifar10/forget_set_1.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:343e9a739404209b65fbb080246b79efb49d938f51a7ab04bc77e0fc8a5890c7
+size 208

forget_sets/cifar10/forget_set_10.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:197fd5153bf80edcc2c48574e11900ab065066f0f8f7608418e05a88a723e33a
+size 928

forget_sets/cifar10/forget_set_2.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:243c897f9120f354248d69fdeafebc4db481b2046b633303c649dff87bb2ff97
+size 928

forget_sets/cifar10/forget_set_3.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef08a53dea26090ec6af7e6c1aeb4f6b00d177b0c04a65c82654ede732ff38ca
+size 8128

forget_sets/cifar10/forget_set_4.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27d014a6eff179de54b8c11d9fe52360a4adc67a58f97ce647d5b8bb5de96a47
+size 208

forget_sets/cifar10/forget_set_5.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3aa610ab3dee7fbaa55afc654b3d76f439197bf83f24cf4100393288bcf8d9a3
+size 928

forget_sets/cifar10/forget_set_6.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08029e0253a1ffc6a435d49e68f824211e125f3b5f4593eddb068131dc20953e
+size 4128

forget_sets/cifar10/forget_set_7.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fae3e72d6c9ee881ac6714544b7079fb9f58c7eb0442196f3891b22e1a0f68e
+size 208

forget_sets/cifar10/forget_set_8.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:521bf3e822a0880ff72d132196ca63f7086d5022788754edc39e1619da3a4f39
+size 928

forget_sets/cifar10/forget_set_9.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05a2e1ad57b8dadf85ffdcd16bae10a4d430d9432426740f772356fef7fdc30c
+size 4128

logits/cifar10/oracle/forget_1/forget/resnet9/id_0_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ca72175cd3f68947225742a0d663d8a02491b2f38d88402324d6a43cdf4b2ec
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_100_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b5a790c4955c7c3012924c87ac04acf37a25cc933dfd86882466e71e9b1b733
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_101_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05da63aa8726f53d32d632d86536242b51e191b85a2be47316abf73eaf1ceaab
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_102_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efdf68186db225710842e802fdf1f67e4cef48a81cb927dee9124923816d8a3e
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_103_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2804a41e4547e0c0d30755f57a3ec147d0a8f1ba730b3d1a575432eec9fb64e
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_104_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b11ed21dfe092c06b14b53866f81c7ba8fb2414c574218dbcb9e473d05cce0fe
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_105_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d8a476b06d607f710b5f6e742ce0464d527fe4ee853982b6bb348468816415d
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_106_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84397f74abb4ea18c77a81b19c87c95b8a23af6630b515dd594be2f3df27c5e2
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_107_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adf135bfa1eb3a3eec9b1510344390ed7b2803db5eab547fb3edb5ff6dc2dbb2
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_108_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d268deabcef492a271f03e63afb070697d440934086008954dbc5c4e365a2ad1
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_109_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c870db8ad7742dbfa6ee7019f7bd2bf2510e07ee8fc90608f176fb357334ffb2
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_10_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11c74647c7d8a4f0eab2985e547bd721e18f606b70c8e46527dfbaf4e29cf0e3
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_110_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f961c64dd41cf14153cdf52b4d191df7c3c459a73e37fbcaa6b00eb372466c0
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_111_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb65a1d93d60155fb869c40c09d9bcdf7cf67aae8d1365b6108d5cd8721b3b82
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_112_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f0e026ec7095fef1ab8f5b9e8ae316964bb41b62a26542bbf17df909ba29fc2
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_113_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14903c356a9d4d46519201e5cdec5909f7da31322b64e7e7e9fa2173e030b8c9
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_114_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4b41f95966bbc1fd807f23405ea69dd9296c0e5352bf855b4808060ece17128
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_115_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d8d740ed27cb6617b28d988573d52b0710e2f71fbda068d9bed7db11baf8d3d
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_116_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b458931860f775b993e453c1ed1fc49625d35d4522b0ecc78d99ad99c1d32419
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_117_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef15a05bb1c2c593031a3a1165710dc21cd916c93d389732d35b2c714a8c537d
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_118_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:632476f2e096cc8b4f5582b2836e60d30d44eb1750c0e96c198879c0bed790e2
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_119_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94f9857c00f81d882d535b426e996ceb783c00992d5820052f0d9f63f459ad52
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_11_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efa189bf2255af7b2a31e01c54abf4032abfef7a77fe13b0dd38976318678d9d
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_120_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0367af152768ec632633bd1867bdfd7f14d45c6b64909f6df8f2068000d5fbb6
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_121_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0181e8c184f7ccde1e2a8c3dac154a3cd9d160f56c87db923af997768619ae0
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_122_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e3065aa4a1fa243e95412a92c90b691a55c9c737670dc3e872807f17afceca1
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_123_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d735c60429a7843e572cd0bde8d45c3fc973a1e76dc73ae29a246d3e0f0b869
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_124_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1578083c893e1eced0ff524b72271da2917466550d477ced14f7e8784ceb88f
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_125_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95105f0e20caa41fcac0f4f25ca697d0d2b425150fc2186f689d3f064f91574d
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_126_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:734696937cc46d2083acf502a1973143767b9bdb841ce1954decd01f8b3c5d71
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_127_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99c5b94d4c037644d9dfe5fdd28ba04dfb7eb58592512c1f32ead1c18ac86728
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_128_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4aa5af9aee7fd50b16d0f9b1d42e50835226662fc5fdedbd9d85ebda72c3940
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_129_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88b6292ff3fd9826086a0daec849a58210ab4a33590c0c40452f28ecb93c9958
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_12_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfe0c1af92c35bb8c33cad348f68aa8d7c909232638444fe633458d4e019b309
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_130_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01560e17c5ed314dd05c0a1f5bacfbe387a04f41490eba0cdab657d5e5d41c4d
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_131_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:282422463c9778fa2594ed8d45993372ceb349a008eb564767ef0b1e47bda98a
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_132_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c85d104c614358a2bb374bfc4ef51716890e7236e1ae82401dece67f406b920
+size 528

logits/cifar10/oracle/forget_1/forget/resnet9/id_133_epoch_23.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cbbf4bd5fa6d00254b43c076e0da77eb269e094a78a34327beb2dc8f48252ac
+size 528