Results 41 to 50 of about 4,436,508 (259)

Fully Variational Noise-Contrastive Estimation

open access: yes, 2023
By using the underlying theory of proper scoring rules, we design a family of noise-contrastive estimation (NCE) methods that are tractable for latent variable models. Both terms in the underlying NCE loss, the one using data samples and the one using noise samples, can be lower-bounded as in variational Bayes, therefore we call this family of losses ...
openaire   +4 more sources

Superimposition of illusory patterns with contrast variations

open access: yesActa Neurobiologiae Experimentalis, 2005
In psychophysical experiments, a perceived length matching task was performed. The stimuli were made of two spatially superimposed illusory figures that differed in structure and luminance contrast but had the same length and coincided precisely, with their ends matched. The contrast of one of the figures was fixed, and that of the other varied.
Bulatov, Aleksandr   +1 more
openaire   +3 more sources

Review: Current progresses of small-angle neutron scattering on soft-matters investigation

open access: yesNuclear Analysis, 2022
A review of the recent works on soft-matters based on small-angle neutron scattering (SANS) is presented. This Review is intended to be a relatively comprehensive source for applying SANS on soft-matters investigation, including the basic principles and ...
Dong Liu   +5 more
doaj   +1 more source

Peeling Plaids Apart: Context Counteracts Cross-Orientation Contrast Masking [PDF]

open access: yes, 2009
Background: Contrast discrimination for an image is usually harder if another image is superimposed on top. We asked whether such contrast masking may be enhanced or relieved depending on cues promoting integration of both images as a single pattern ...
Verghese, P.   +7 more
core   +2 more sources

The contrast effect in a competency based situational interview [PDF]

open access: yes, 2004
The recruitment interview is used ubiquitously by organisations in the UK as part of the recruitment and selection process. Despite improvements over the years, the method is still prone to error and it is important for organisations to take what steps
Mills, Timothy
core   +3 more sources

Variational Supervised Contrastive Learning

open access: yesAdvances in Neural Information Processing Systems 38
Contrastive learning has proven to be highly efficient and adaptable in shaping representation spaces across diverse modalities by pulling similar samples together and pushing dissimilar ones apart. However, two key limitations persist: (1) Without explicit regulation of the embedding distribution, semantically related instances can inadvertently be ...
Ziwen Wang   +4 more
openaire   +2 more sources

Central Nervous System Tumors Among Infants in Canada: A Report From CYP‐C

open access: yesPediatric Blood &Cancer, EarlyView.
ABSTRACT Background Central nervous system (CNS) tumors in infants are rare, pose unique clinical challenges, and lack large‐scale evidence‐based data to guide management. This study seeks to describe CNS tumors in Canadian infants and to compare their outcomes with those of older children.
Samuel Sassine   +17 more
wiley   +1 more source

Alignment of contrast enhanced medical images [PDF]

open access: yes, 2009
The re-alignment of series of medical images in which there are multiple contrast variations is difficult. The reason for this is that the popularmeasures of image similarity used to drive the alignment procedure do not separate the influence of ...
Melbourne, A.
core  

Contrast threshold elevation following continuous and interrupted adaptation [PDF]

open access: yes, 1986
Contrast thresholds for a 6 c/deg sinewave grating were measured following continuous and interrupted adaptation of 10 min duration to a high-contrast (0.6) sinewave grating of the same spatial frequency. Interrupted adaptation was administered as five 2-
Magnussen, S., Greenlee, Mark W.
core   +1 more source

Probabilistic Variational Contrastive Learning

open access: yesCoRR
Deterministic embeddings learned by contrastive learning (CL) methods such as SimCLR and SupCon achieve state-of-the-art performance but lack a principled mechanism for uncertainty quantification. We propose Variational Contrastive Learning (VCL), a decoder-free framework that maximizes the evidence lower bound (ELBO) by interpreting the InfoNCE loss ...
Minoh Jeong, Seonho Kim, Alfred Hero
openaire   +2 more sources

Home - About - Disclaimer - Privacy