SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.03295 · 2023

Can pre-trained models assist in dataset distillation?

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 8 functions out of this paper's own repositories and ran 2 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.

RepositoryRoleRan
yaolu-zjut/ddinterpreter canonical 2 of 8
FunctionStatusWhere it lives
distance_wb Ran yaolu-zjut/ddinterpreter/methods/DC_DSA_DM/utils.py
pointer only (licence: NONE) · get_code("6e9c65aaa7c1b5da")
get_dataset_info Ran yaolu-zjut/ddinterpreter/utils_clom/Dataloader.py
pointer only (licence: NONE) · get_code("a2c08a19541743bb")
VGG11 Not yet run yaolu-zjut/ddinterpreter/utils_clom/model.py
pointer only (licence: NONE) · get_code("8e1b179ef1f0fe03")
VGG11BN Not yet run yaolu-zjut/ddinterpreter/utils_clom/model.py
pointer only (licence: NONE) · get_code("cfab0fafbf988aa4")
VGG13 Not yet run yaolu-zjut/ddinterpreter/utils_clom/model.py
pointer only (licence: NONE) · get_code("07962bf28fe347a0")
get_dataset Not yet run yaolu-zjut/ddinterpreter/methods/DC_DSA_DM/utils.py
pointer only (licence: NONE) · get_code("4ede4e38dc61074f")
get_dataset Not yet run yaolu-zjut/ddinterpreter/utils_clom/Dataloader.py
pointer only (licence: NONE) · get_code("d6e228049010aa8c")
get_network Not yet run yaolu-zjut/ddinterpreter/utils_clom/model_pool.py
pointer only (licence: NONE) · get_code("ef0e8481768a2222")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Dataset Distillation (DD) is a prominent technique that encapsulates knowledge from a large-scale original dataset into a small synthetic dataset for efficient training. Meanwhile, Pre-trained Models (PTMs) function as knowledge repositories, containing extensive information from the original dataset. This naturally raises a question: Can PTMs effectively transfer knowledge to synthetic datasets, guiding DD accurately? To this end, we conduct preliminary experiments, confirming the contribution of PTMs to DD. Afterwards, we systematically study different options in PTMs, including initialization parameters, model architecture, training epoch and domain knowledge, revealing that: 1) Increasing model diversity enhances the performance of synthetic datasets; 2) Sub-optimal models can also assist in DD and outperform well-trained ones in certain cases; 3) Domain-specific PTMs are not mandatory for DD, but a reasonable domain match is crucial. Finally, by selecting optimal options, we significantly improve the cross-architecture generalization over baseline DD methods. We hope our work will facilitate researchers to develop better DD techniques. Our code is available at https://github.com/yaolu-zjut/DDInterpreter.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2310.03295")
get_code_for_paper("2310.03295")
have("2310.03295")

Connect an agent — have() is free.