This Supplement contains additional materials related to the paper PUnDA: Probabilistic Unsupervised Domain Adaptation for Knowledge Transfer Across Visual Categories. In particular, in Sec. 2 we present additional results that qualitatively demonstrate the advantages of our adaptation framework over competing approaches such as the ILS [2]. We contrast 2D embeddings of the features adapted by ...

The paper describes the process of program comprehension from the point of view of constructivist theory of learning. According to this view, program comprehension starts with the pre-existing knowledge and continues through processes of assimilation and adaptation. The assimilation means that the facts encountered in the program are either added to the knowledge or rejected. Adaptation means t...

INTRODUCTION To determine whether foveal light and dark adaptation are affected in glaucoma. METHODS Case-control study with 23 glaucoma patients and 51 controls. Light and dark adaptation were measured twice. After 10 minutes pre-adaptation to 0.0032 cd/m2, the background luminance increased stepwise to 320 (5 log unit step) or 10,000 cd/m2 (6.5 log unit step) for 10 minutes, then it decreas...

Plants produce unique natural products as a result of gene mutation and subsequent adaptation of metabolic pathways to create new secondary metabolites. However, their biosynthesis and accumulation remains remarkably under the control of the biotic and abiotic environments. Alkaloid biosynthesis, which requires the adaptation of cellular activities to perform specialized metabolism without comp...

In this work, we propose a semisupervised extension to a well-known supervised domain adaptation approach (EA) (Daumé III, 2007). Our proposed approach (EA++) builds on the notion of augmented space (introduced in EA) and harnesses unlabeled data in target domain to ameliorate the transfer of information from source to target. This semisupervised approach to domain adaptation is extremely simpl...

An anticipatory system S2 is one which contains a model of a system S1 with which it interacts. This model is a predictive model; its present states provide information about future states of S1. Further, the present state of the model causes a change of state in other subsystems of S2; these subsystems are (a) involved in the interaction of S2 with S1, and (b) they do not affect (that is, are ...

We address the problem of weakly supervised object localization where only image-level annotations are available for training object detectors. Numerous methods have been proposed to tackle this problem through mining object proposals. However, a substantial amount of noise in object proposals causes ambiguities for learning discriminative object models. Such approaches are sensitive to model i...

The processing of orientations is at the core of our visual experience. Orientation selectivity in human visual cortex has been inferred from psychophysical experiments and more recently demonstrated with functional magnetic resonance imaging (fMRI). One method to identify orientation-selective responses is fMRI adaptation, in which two stimuli-either with the same or with different orientation...

Language Models (LMs) for Automatic Speech Recognition (ASR) are typically trained on large text corpora from news articles, books and web documents. These types of corpora, however, are unlikely to match the test distribution of ASR systems, which expect spoken utterances. Therefore, the LM is typically adapted to a smaller held-out in-domain dataset that is drawn from the test distribution. W...

The forthcoming wireless systems beyond 3G are heralded to trigger dramatic changes in the way that mobile services are being developed and delivered. User expectations are expected to raise to a significantly higher level, towards the demand for terminal-, networkand location-aware provision of ubiquitous, personalized multimedia services. Adaptability, a feature that did not attract particula...