Source-free domain adaptation (SFDA) aims to adapt a classifier to an unlabelled target data set by only using a pre-trained source model. However, the absence of the source data and the domain shift makes the predictions on the target data unreliable. We propose quantifying the uncertainty in the source model predictions and utilizing it to guide the target adaptation. For this, we construct a probabilistic source model by incorporating priors on the network parameters inducing a distribution over the model predictions. Uncertainties are estimated by employing a Laplace approximation and incorporated to identify target data points that do not lie in the source manifold and to down-weight them when maximizing the mutual information on the target data. Unlike recent works, our probabilistic treatment is computationally lightweight, decouples source training and target adaptation, and requires no specialized source training or changes of the model architecture. We show the advantages of uncertainty-guided SFDA over traditional SFDA in the closed-set and open-set settings and provide empirical evidence that our approach is more robust to strong domain shifts even without tuning.
Three fluorescent probes were utilized to investigate the microenvironment created by a watersoluble polyphosphazene copolymer in aqueous solution. The combination of intensity, wavelength shift, and vibrational analysis measurements of the three small-molecule probes is used to postulate an environment at high pH (with no added electrolyte) that is similar in polarity to methanol. In the absence of excess small-molecule electrolyte, negatively charged probes are repelled by the charge on the polymer. Addition of small-molecule electrolyte results in solubilization of the negatively charged probes. However, pyrene, even in the absence of added electrolyte (NaCl), is solubilized. Investigation of pyrene fluorescence quenching demonstrates the ability of the polyphosphazene copolymer to bind positively charged ions (T1+) and repel negatively charged ions (I-). Neutral quenchers such as nitromethane readily penetrate the microenvironment of the polymer.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.