Abstract
Bioacoustic call-type classification relies on costly expert annotation. Active learning can reduce this burden by selecting a small batch of segments for expert annotation and using the labeled segments for training the classifier. The setting is hard: the target calls are extremely sparse and the call-type distribution is long-tailed, so a tight budget must be spent on the few rare, informative segments.
We propose BADGE-Greedy-DPP, a deterministic batch selector that greedily adds the segment whose BADGE gradient embedding most enlarges the volume spanned by the batch; because this log-volume objective is submodular, the greedy rule guarantees a batch value at least a $(1-1/e)$ fraction of the optimum of this objective, a guarantee not provided by BADGE's existing k-means++ and MCMC DPP sampling heuristics.
There is also a temporal granularity mismatch in the task. The acquisition function scores whole segments, yet the informative frames inside them are few. Uniform averaging therefore washes them out. We show that the BADGE construction naturally addresses this mismatch when applied frame-wise, as prediction residuals weight the aggregated pseudo-gradient, so confidently predicted no-call frames contribute little while a single uncertain rare-call frame can still set the segment's direction.
Across 10 runs on a sparse, imbalanced hyena call-type dataset, BADGE-Greedy-DPP achieves the best overall and rare-call-type performance among all compared query strategies, including MFFT, the strongest non-BADGE baseline, and the two vanilla BADGE traversals.
Blogger's Review: The BADGE-Greedy-DPP method proposed in this paper shows remarkable advantages in handling bioacoustic active learning under long-tailed distributions, especially with limited budgets. The greedy strategy effectively enhances the selection efficiency of rare calls, which is highly practical and has great potential for broader applications. Its solution to the temporal granularity mismatch also provides new insights for related research fields.