Showing papers by "Andrei Rusu published in 2020"
TL;DR: This review relates continual learning to the learning dynamics of neural networks, highlighting the potential it has to considerably improve data efficiency and consider the many new biologically inspired approaches that have emerged in recent years.
222 citations
Proceedings Article•
30 Apr 2020
TL;DR: WarpGrad as mentioned in this paper meta-learns an efficiently parameterized preconditioning matrix that facilitates gradient descent across the task distribution by interleaving nonlinear layers, referred to as warp-layers, between the layers of a task-learner.
Abstract: Learning an efficient update rule from data that promotes rapid learning of new tasks from the same distribution remains an open problem in meta-learning. Typically, previous works have approached this issue either by attempting to train a neural network that directly produces updates or by attempting to learn better initialisations or scaling factors for a gradient-based update rule. Both these approaches pose challenges. On one hand, directly producing an update forgoes a useful inductive bias and can easily lead to non-converging behaviour. On the other hand, approaches that try to control a gradient-based update rule typically resort to computing gradients through the learning process to obtain their meta-gradients, leading to methods that can not scale beyond few-shot task adaptation. In this work we propose Warped Gradient Descent (WarpGrad), a method that intersects these approaches to mitigate their limitations. WarpGrad meta-learns an efficiently parameterised preconditioning matrix that facilitates gradient descent across the task distribution. Preconditioning arises by interleaving non-linear layers, referred to as warp-layers, between the layers of a task-learner. Warp-layers are meta-learned without backpropagating through the task training process in a manner similar to methods that learn to directly produce updates. WarpGrad is computationally efficient, easy to implement, and can scale to arbitrarily large meta-learning problems. We provide a geometrical interpretation of the approach and evaluate its effectiveness in a variety of settings, including few-shot, standard supervised, continual and reinforcement learning.
76 citations
TL;DR: In this paper, the authors proposed a model of personality characteristics (i.e., proactive personality, core self-evaluation, and psychological capital) which can directly predict work engagement, and indirectly, employees' job performance and their mental health.
60 citations
51 citations
TL;DR: In this paper, a dynamic and multi-level perspective on identification with the group was explored, and the extent to which core self-evaluations, study engagement, group development, and relationship conflict influence identification with groups was explored.
9 citations
TL;DR: This paper explored implicit attitudes towards Catalan and Spanish in the multicultural and multilingual context in the context of implicit language attitudes, and found that implicit attitudes toward Spanish and Catalan are related to each other.
Abstract: In line with the increased interest in studying implicit language attitudes, this study aims to explore implicit attitudes towards Catalan and Spanish in the multicultural and multilingual context ...
7 citations
TL;DR: All studies found that participants often failed to use the skip option to exert control over conditioned preferences, and even when explicitly warned that the pairings could influence them, participants seemed to believe that they were not vulnerable to such effects.
Abstract: Evaluative conditioning procedures change people’s evaluations of stimuli that are paired with pleasant or unpleasant items. To test whether influence awareness allows people to resist such persuas...
2 citations
01 Sep 2020
TL;DR: In this article, the color characteristics of the Facebook profile picture of 508 Romanian users were used to predict the level of neuroticism on a sample of 5008 users, and four machine learning algorithms were evaluated.
Abstract: Previous research has mostly focused on the link between the linguistic and behavioral footprints found on social media on the one hand and personality on the other. Despite the high amount of image-based contents posted and shared online and the valuable implicit information they might conceal about users' preferences and tendencies, the study of visual traces is in its infancy. The goal of the current paper is to test whether the color characteristics of the Facebook profile picture could mirror the level of neuroticism on a sample of 508 Romanian users. For this purpose, we assessed the classification performance of four machine learning algorithms having as input three sets of visual features: (1) the colorfulness, which indicates how colorful is an image; (2) the proportion of cold colors, along with the mean and standard deviation for saturation and value; (3) the emotional load, defined as pleasure, arousal, and dominance. None of the models showed good accuracy. However, this paper contributes to the literature by being part of a line of research that requires development not only in Romania but also worldwide.
1 citations
TL;DR: In this paper, the main and interaction effects of membership change and normative interventions aimed at increasing collaboration and participation on cognitive decline in group cognitive decline were tested in an experimental study, where participants were asked to participate in group activities.
Abstract: In this experimental study, we tested main and interaction effects of membership change and normative interventions aimed at increasing collaboration and participation on cognitive decline in group...
1 citations