SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.03885 · 2022

Meta-DMoE: Adapting to Domain Shift by Meta-Distillation from Mixture-of-Experts

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 7 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.

RepositoryRoleRan
n3il666/meta-dmoe canonical 4 of 7
FunctionStatusWhere it lives
get_model_accuracy Ran n3il666/meta-dmoe/src/camelyon/camelyon_experts.py
code served (permissive licence) · get_code("f6e35c72666b64f6")
get_selector_accuracy Ran n3il666/meta-dmoe/src/fmow/fmow_aggregator.py
code served (permissive licence) · get_code("e2705ed6b48d20e5")
initialize_image_base_transform Ran n3il666/meta-dmoe/src/camelyon/camelyon_utils.py
code served (permissive licence) · get_code("48d7a611a78faa01")
initialize_image_base_transform Ran n3il666/meta-dmoe/src/fmow/fmow_utils.py
code served (permissive licence) · get_code("69e94cb234255b0c")
get_expert_split Not yet run n3il666/meta-dmoe/src/camelyon/camelyon_experts.py
code served (permissive licence) · get_code("a952a1fa7c7d8616")
get_model_accuracy Not yet run n3il666/meta-dmoe/src/fmow/fmow_experts.py
code served (permissive licence) · get_code("f2746799d8f51cf0")
get_selector_accuracy Not yet run n3il666/meta-dmoe/src/camelyon/camelyon_aggregator.py
code served (permissive licence) · get_code("c71f59f1661f12e4")

Repositories linked to this paper

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

Abstract

In this paper, we tackle the problem of domain shift. Most existing methods perform training on multiple source domains using a single model, and the same trained model is used on all unseen target domains. Such solutions are sub-optimal as each target domain exhibits its own specialty, which is not adapted. Furthermore, expecting single-model training to learn extensive knowledge from multiple source domains is counterintuitive. The model is more biased toward learning only domain-invariant features and may result in negative knowledge transfer. In this work, we propose a novel framework for unsupervised test-time adaptation, which is formulated as a knowledge distillation process to address domain shift. Specifically, we incorporate Mixture-of-Experts (MoE) as teachers, where each expert is separately trained on different source domains to maximize their specialty. Given a test-time target domain, a small set of unlabeled data is sampled to query the knowledge from MoE. As the source domains are correlated to the target domains, a transformer-based aggregator then combines the domain knowledge by examining the interconnection among them. The output is treated as a supervision signal to adapt a student prediction network toward the target domain. We further employ meta-learning to enforce the aggregator to distill positive knowledge and the student network to achieve fast adaptation. Extensive experiments demonstrate that the proposed method outperforms the state-of-the-art and validates the effectiveness of each proposed component. Our code is available at https://github.com/n3il666/Meta-DMoE.

For agents

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

get_harvested_code_for_paper("2210.03885")
get_code_for_paper("2210.03885")
have("2210.03885")

Connect an agent — have() is free.