REVIEW 13 cited by
Domain Adaptation: Learning Bounds and Algorithms
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
This paper addresses the general problem of domain adaptation which arises in a variety of applications where the distribution of the labeled sample available somewhat differs from that of the test data. Building on previous work by Ben-David et al. (2007), we introduce a novel distance between distributions, discrepancy distance, that is tailored to adaptation problems with arbitrary loss functions. We give Rademacher complexity bounds for estimating the discrepancy distance from finite samples for different loss functions. Using this distance, we derive novel generalization bounds for domain adaptation for a wide family of loss functions. We also present a series of novel adaptation bounds for large classes of regularization-based algorithms, including support vector machines and kernel ridge regression based on the empirical discrepancy. This motivates our analysis of the problem of minimizing the empirical discrepancy for various loss functions for which we also give novel algorithms. We report the results of preliminary experiments that demonstrate the benefits of our discrepancy minimization algorithms for domain adaptation.
Forward citations
Cited by 13 Pith papers
-
Mixed-Sample SGD: an End-to-end Analysis of Supervised Transfer Learning
An adaptive mixed-sample SGD procedure is claimed to converge at 1/sqrt(T) to a solution whose target risk matches the better of source-only and target-only ERM, but the main convergence bound contains a non-vanishing...
-
Phase Transition in Nonparametric Minimax Rates for Covariate Shifts on Approximate Manifolds
Under covariate shift with target data near a smooth d-dimensional manifold in D dimensions, the minimax regression rate switches between a manifold-dominated and a noise-dominated regime at a threshold set by source ...
-
Clustered Federated Learning via Embedding Distributions
EMD-CFL clusters federated learning clients in one shot by comparing Earth Mover's distances between randomly projected embedding distributions, matching oracle clustering on several benchmarks.
-
Subgroups Matter for Robust Bias Mitigation
Subgroup choice determines whether bias mitigation helps or hurts, and the minimum KL divergence to the unbiased test distribution predicts success.
-
Nonparametric Goodness-of-fit Testing under Covariate Shift
Truncated importance-weighted kernel ridge regression with multiplier bootstrap yields valid L2(Q) confidence balls for nonparametric goodness-of-fit under covariate shift.
-
ACE and Diverse Generalization via Selective Disagreement
ACE learns an ensemble of classifiers that agree on labeled data but confidently and selectively disagree on target-distribution data, recovering diverse human-interpretable concepts under complete spurious correlation.
-
Enhancing Fatigue Detection through Heterogeneous Multi-Source Data Integration and Cross-Domain Modality Imputation
A framework that imputes missing high-fidelity sensor channels (EEG, ECG) from shared wearable signals improves fatigue detection accuracy in a sensor-constrained target dataset in several configurations.
-
Topology-Aware Modeling for Unsupervised Simulation-to-Reality Point Cloud Recognition
A topology-aware self-supervised framework improves unsupervised simulation-to-reality point cloud classification by combining Fourier-encoded global structure, local implicit fields, and contrastive self-training.
-
Best Arm Identification with Possibly Biased Offline Data
LUCB-H adaptively combines offline and online data for best arm identification, matching or beating standard LUCB depending on whether the historical data is helpful or misleading.
-
Measuring Representational Shifts in Continual Learning: A Linear Transformation Perspective
Representation discrepancy, a new metric with theoretical bounds, shows continual learning forgets features faster in deeper layers and slower in wider networks.
-
Computational Efficiency under Covariate Shift in Kernel Ridge Regression
Nyström-approximated importance-weighted kernel ridge regression achieves the same optimal excess-risk rates as the full method under covariate shift, with sublinear time and memory costs.
-
On the Transferability and Discriminability of Repersentation Learning in Unsupervised Domain Adaptation
RLGLC combines a relaxed Wasserstein alignment with a contrastive local consistency term for UDA and reports SOTA results, but the proof that such a term is necessary is not rigorous.
-
YOLOv14:Unified Cross-Domain Real-Time Object Detectionwith Adaptive Multi-View Representation
YOLOv14 reports a real-time detector with 49.1 COCO mAP and cross-domain gains, but its headline game-character gain is measured on a benchmark synthesized from its own training augmentation.
Discussion (0). Continue with ORCID to comment.