Diversity-aware Buffer for Coping with Temporally Correlated Data Streams in Online Test-time Adaptation
This addresses performance drops in deployed models due to non-i.i.d. data streams, offering a stable adaptation method, though it is incremental as it builds on existing TTA approaches.
The paper tackles the problem of online test-time adaptation (TTA) for models facing temporally correlated, non-i.i.d. data streams, which degrade performance, by proposing a diversity-aware buffer to simulate i.i.d. streams and a weighted loss, achieving state-of-the-art results on ImageNet benchmarks.
Since distribution shifts are likely to occur after a model's deployment and can drastically decrease the model's performance, online test-time adaptation (TTA) continues to update the model during test-time, leveraging the current test data. In real-world scenarios, test data streams are not always independent and identically distributed (i.i.d.). Instead, they are frequently temporally correlated, making them non-i.i.d. Many existing methods struggle to cope with this scenario. In response, we propose a diversity-aware and category-balanced buffer that can simulate an i.i.d. data stream, even in non-i.i.d. scenarios. Combined with a diversity and entropy-weighted entropy loss, we show that a stable adaptation is possible on a wide range of corruptions and natural domain shifts, based on ImageNet. We achieve state-of-the-art results on most considered benchmarks.