The Experts below are selected from a list of 321 Experts worldwide ranked by ideXlab platform
Genevieve A. Dingle - One of the best experts on this subject based on the ideXlab platform.
-
social Identity Mapping in addiction recovery sim ar extension and application of a visual method
Addiction Research & Theory, 2019Co-Authors: Melinda Beckwith, Catherine Haslam, Genevieve A. Dingle, David Best, Jock Mackenzie, Michael Savic, Ramez Bathish, Petra K Staiger, Dan I LubmanAbstract:Background: The Social Identity approach offers a unifying framework for understanding recovery from addiction as a process of Identity change, associated with change in social network composition....
-
Social Identity Mapping: measuring social Identity change in recovery from addiction
2016Co-Authors: Catherine Haslam, Genevieve A. Dingle, David Best, Jock Mackenzie, Melinda BeckwithAbstract:Throughout our lives periods of change and transition are inevitable, whether it is starting school, taking up employment, moving house, having children, retiring, or moving into care. Some transitions are planned and some unexpected, but all are associated with varying degrees of uncertainty; even those that change our lives for the better. How we manage that uncertainty is key to protecting well-being and ensuring successful transition, and this, we argue, has its basis in the availability of strong supportive networks of relationships with others, particularly groups of others. It is from these relationships that we gain the support required to manage and adjust to change, and this is even more vital when that change is especially challenging, as in the case of recovery from addiction. As we will argue, the challenge in this life transition lies in engineering the process of social Identity change – one that involves moving away from drug using networks to sustained engagement with networks supportive of recovery. Social Identity theorizing provides a framework to understand the role that social identities play in this transition and a basis from which to map their influence. In this chapter, we outline a theoretical framework that explains adjustment to social Identity change and introduce the Social Identity Mapping (SIM) tool to measure such change in the process of recovery.
-
Social Identity Mapping: A procedure for visual representation and assessment of subjective multiple group memberships.
The British journal of social psychology, 2016Co-Authors: Tegan Cruwys, Catherine Haslam, Niklas K. Steffens, S. Alexander Haslam, Jolanda Jetten, Genevieve A. DingleAbstract:In this research, we introduce Social Identity Mapping (SIM) as a method for visually representing and assessing a person's subjective network of group memberships. To provide evidence of its utility, we report validating data from three studies (two longitudinal), involving student, community, and clinical samples, together comprising over 400 participants. Results indicate that SIM is easy to use, internally consistent, with good convergent and discriminant validity. Each study also illustrates the ways that SIM can be used to address a range of novel research questions. Study 1 shows that multiple positive group memberships are a particularly powerful predictor of well-being. Study 2 shows that social support is primarily given and received within social groups and that only in-group support is beneficial for well-being. Study 3 shows that improved mental health following a social group intervention is attributable to an increase in group compatibility. In this way, the studies demonstrate the capacity for SIM to make a contribution both to the development of social-psychological theory and to its practical application.
Clinton Fookes - One of the best experts on this subject based on the ideXlab platform.
-
semantic consistency and Identity Mapping multi component generative adversarial network for person re identification
Workshop on Applications of Computer Vision, 2020Co-Authors: Amena Khatun, Simon Denman, Sridha Sridharan, Clinton FookesAbstract:In a real world environment, person re-identification (Re-ID) is a challenging task due to variations in lighting conditions, viewing angles, pose and occlusions. Despite recent performance gains, current person Re-ID algorithms still suffer heavily when encountering these variations. To address this problem, we propose a semantic consistency and Identity Mapping multi-component generative adversarial network (SC-IMGAN) which provides style adaptation from one to many domains. To ensure that transformed images are as realistic as possible, we propose novel Identity Mapping and semantic consistency losses to maintain Identity across the diverse domains. For the Re-ID task, we propose a joint verification-identification quartet network which is trained with generated and real images, followed by an effective quartet loss for verification. Our proposed method outperforms state-of-the-art techniques on six challenging person Re-ID datasets: CUHK01, CUHK03, VIPeR, PRID2011, iLIDS and Market-1501.
-
WACV - Semantic Consistency and Identity Mapping Multi-Component Generative Adversarial Network for Person Re-Identification
2020 IEEE Winter Conference on Applications of Computer Vision (WACV), 2020Co-Authors: Amena Khatun, Simon Denman, Sridha Sridharan, Clinton FookesAbstract:In a real world environment, person re-identification (Re-ID) is a challenging task due to variations in lighting conditions, viewing angles, pose and occlusions. Despite recent performance gains, current person Re-ID algorithms still suffer heavily when encountering these variations. To address this problem, we propose a semantic consistency and Identity Mapping multi-component generative adversarial network (SC-IMGAN) which provides style adaptation from one to many domains. To ensure that transformed images are as realistic as possible, we propose novel Identity Mapping and semantic consistency losses to maintain Identity across the diverse domains. For the Re-ID task, we propose a joint verification-identification quartet network which is trained with generated and real images, followed by an effective quartet loss for verification. Our proposed method outperforms state-of-the-art techniques on six challenging person Re-ID datasets: CUHK01, CUHK03, VIPeR, PRID2011, iLIDS and Market-1501.
Catherine Haslam - One of the best experts on this subject based on the ideXlab platform.
-
social Identity Mapping in addiction recovery sim ar extension and application of a visual method
Addiction Research & Theory, 2019Co-Authors: Melinda Beckwith, Catherine Haslam, Genevieve A. Dingle, David Best, Jock Mackenzie, Michael Savic, Ramez Bathish, Petra K Staiger, Dan I LubmanAbstract:Background: The Social Identity approach offers a unifying framework for understanding recovery from addiction as a process of Identity change, associated with change in social network composition....
-
Social Identity Mapping: measuring social Identity change in recovery from addiction
2016Co-Authors: Catherine Haslam, Genevieve A. Dingle, David Best, Jock Mackenzie, Melinda BeckwithAbstract:Throughout our lives periods of change and transition are inevitable, whether it is starting school, taking up employment, moving house, having children, retiring, or moving into care. Some transitions are planned and some unexpected, but all are associated with varying degrees of uncertainty; even those that change our lives for the better. How we manage that uncertainty is key to protecting well-being and ensuring successful transition, and this, we argue, has its basis in the availability of strong supportive networks of relationships with others, particularly groups of others. It is from these relationships that we gain the support required to manage and adjust to change, and this is even more vital when that change is especially challenging, as in the case of recovery from addiction. As we will argue, the challenge in this life transition lies in engineering the process of social Identity change – one that involves moving away from drug using networks to sustained engagement with networks supportive of recovery. Social Identity theorizing provides a framework to understand the role that social identities play in this transition and a basis from which to map their influence. In this chapter, we outline a theoretical framework that explains adjustment to social Identity change and introduce the Social Identity Mapping (SIM) tool to measure such change in the process of recovery.
-
Social Identity Mapping: A procedure for visual representation and assessment of subjective multiple group memberships.
The British journal of social psychology, 2016Co-Authors: Tegan Cruwys, Catherine Haslam, Niklas K. Steffens, S. Alexander Haslam, Jolanda Jetten, Genevieve A. DingleAbstract:In this research, we introduce Social Identity Mapping (SIM) as a method for visually representing and assessing a person's subjective network of group memberships. To provide evidence of its utility, we report validating data from three studies (two longitudinal), involving student, community, and clinical samples, together comprising over 400 participants. Results indicate that SIM is easy to use, internally consistent, with good convergent and discriminant validity. Each study also illustrates the ways that SIM can be used to address a range of novel research questions. Study 1 shows that multiple positive group memberships are a particularly powerful predictor of well-being. Study 2 shows that social support is primarily given and received within social groups and that only in-group support is beneficial for well-being. Study 3 shows that improved mental health following a social group intervention is attributable to an increase in group compatibility. In this way, the studies demonstrate the capacity for SIM to make a contribution both to the development of social-psychological theory and to its practical application.
Amena Khatun - One of the best experts on this subject based on the ideXlab platform.
-
semantic consistency and Identity Mapping multi component generative adversarial network for person re identification
Workshop on Applications of Computer Vision, 2020Co-Authors: Amena Khatun, Simon Denman, Sridha Sridharan, Clinton FookesAbstract:In a real world environment, person re-identification (Re-ID) is a challenging task due to variations in lighting conditions, viewing angles, pose and occlusions. Despite recent performance gains, current person Re-ID algorithms still suffer heavily when encountering these variations. To address this problem, we propose a semantic consistency and Identity Mapping multi-component generative adversarial network (SC-IMGAN) which provides style adaptation from one to many domains. To ensure that transformed images are as realistic as possible, we propose novel Identity Mapping and semantic consistency losses to maintain Identity across the diverse domains. For the Re-ID task, we propose a joint verification-identification quartet network which is trained with generated and real images, followed by an effective quartet loss for verification. Our proposed method outperforms state-of-the-art techniques on six challenging person Re-ID datasets: CUHK01, CUHK03, VIPeR, PRID2011, iLIDS and Market-1501.
-
WACV - Semantic Consistency and Identity Mapping Multi-Component Generative Adversarial Network for Person Re-Identification
2020 IEEE Winter Conference on Applications of Computer Vision (WACV), 2020Co-Authors: Amena Khatun, Simon Denman, Sridha Sridharan, Clinton FookesAbstract:In a real world environment, person re-identification (Re-ID) is a challenging task due to variations in lighting conditions, viewing angles, pose and occlusions. Despite recent performance gains, current person Re-ID algorithms still suffer heavily when encountering these variations. To address this problem, we propose a semantic consistency and Identity Mapping multi-component generative adversarial network (SC-IMGAN) which provides style adaptation from one to many domains. To ensure that transformed images are as realistic as possible, we propose novel Identity Mapping and semantic consistency losses to maintain Identity across the diverse domains. For the Re-ID task, we propose a joint verification-identification quartet network which is trained with generated and real images, followed by an effective quartet loss for verification. Our proposed method outperforms state-of-the-art techniques on six challenging person Re-ID datasets: CUHK01, CUHK03, VIPeR, PRID2011, iLIDS and Market-1501.
Yang Yuan - One of the best experts on this subject based on the ideXlab platform.
-
NIPS - Convergence Analysis of Two-layer Neural Networks with ReLU Activation
2017Co-Authors: Yang YuanAbstract:In recent years, stochastic gradient descent (SGD) based techniques has become the standard tools for training neural networks. However, formal theoretical understanding of why SGD can train neural networks in practice is largely missing. In this paper, we make progress on understanding this mystery by providing a convergence analysis for SGD on a rich subset of two-layer feedforward networks with ReLU activations. This subset is characterized by a special structure called "Identity Mapping". We prove that, if input follows from Gaussian distribution, with standard $O(1/\sqrt{d})$ initialization of the weights, SGD converges to the global minimum in polynomial number of steps. Unlike normal vanilla networks, the "Identity Mapping" makes our network asymmetric and thus the global minimum is unique. To complement our theory, we are also able to show experimentally that multi-layer networks with this Mapping have better performance compared with normal vanilla networks. Our convergence theorem differs from traditional non-convex optimization techniques. We show that SGD converges to optimal in "two phases": In phase I, the gradient points to the wrong direction, however, a potential function $g$ gradually decreases. Then in phase II, SGD enters a nice one point convex region and converges. We also show that the Identity Mapping is necessary for convergence, as it moves the initial point to a better place for optimization. Experiment verifies our claims.
-
convergence analysis of two layer neural networks with relu activation
Neural Information Processing Systems, 2017Co-Authors: Yuanzhi Li, Yang YuanAbstract:In recent years, stochastic gradient descent (SGD) based techniques has become the standard tools for training neural networks. However, formal theoretical understanding of why SGD can train neural networks in practice is largely missing. In this paper, we make progress on understanding this mystery by providing a convergence analysis for SGD on a rich subset of two-layer feedforward networks with ReLU activations. This subset is characterized by a special structure called "Identity Mapping". We prove that, if input follows from Gaussian distribution, with standard $O(1/\sqrt{d})$ initialization of the weights, SGD converges to the global minimum in polynomial number of steps. Unlike normal vanilla networks, the "Identity Mapping" makes our network asymmetric and thus the global minimum is unique. To complement our theory, we are also able to show experimentally that multi-layer networks with this Mapping have better performance compared with normal vanilla networks. Our convergence theorem differs from traditional non-convex optimization techniques. We show that SGD converges to optimal in "two phases": In phase I, the gradient points to the wrong direction, however, a potential function $g$ gradually decreases. Then in phase II, SGD enters a nice one point convex region and converges. We also show that the Identity Mapping is necessary for convergence, as it moves the initial point to a better place for optimization. Experiment verifies our claims.
