We study joint learning of network topology and a mixed opinion dynamics, in which agents may have different update rules. Such a model captures the diversity of real individual interactions. We propose a learning alg...
详细信息
Fairness has become an important concern in Federated learning (FL). An unfair model that performs well for some clients while performing poorly for others can reduce the willingness of clients to participate. In this...
详细信息
ISBN:
(纸本)1577358872
Fairness has become an important concern in Federated learning (FL). An unfair model that performs well for some clients while performing poorly for others can reduce the willingness of clients to participate. In this work, we identify a direct cause of unfairness in FL - the use of an unfair direction to update the global model, which favors some clients while conflicting with other clients' gradients at the model and layer levels. To address these issues, we propose a layer-wise fair Federated learning algorithm (FedLF). Firstly, we formulate a multi-objective optimization problem with an effective fair-driven objective for FL. A layer-wise fair direction is then calculated to mitigate the model and layer-level gradient conflicts and reduce the improvement bias. We further provide the theoretical analysis on how FedLF can improve fairness and guarantee convergence. Extensive experiments on different learning tasks and models demonstrate that FedLF outperforms the SOTA FL algorithms in terms of accuracy and fairness. The source code is available at https://***/zibinpan/FedLF.
Humans and animals develop learning-to-learn strategies throughout their lives to accelerate learning. One theory suggests that this is achieved by a metacognitive process of controlling and monitoring learning. Altho...
详细信息
Humans and animals develop learning-to-learn strategies throughout their lives to accelerate learning. One theory suggests that this is achieved by a metacognitive process of controlling and monitoring learning. Although such learning-to-learn is also observed in motor learning, the metacognitive aspect of learning regulation has not been considered in classical theories of motor learning. Here, we formulated a minimal mechanism of this process as reinforcement learning of motor learning properties, which regulates a policy for memory update in response to sensory prediction error while monitoring its performance. This theory was confirmed in human motor learning experiments, in which the subjective sense of learning-outcome association determined the direction of up- and down-regulation of both learning speed and memory retention. Thus, it provides a simple, unifying account for variations in learning speeds, where the reinforcement learning mechanism monitors and controls the motor learning process. Metacognition is fundamental for regulating learning speeds and memory retention. Here, the authors demonstrate that reinforcement learning mediates this process in implicit motor learning, maximizing rewards and minimizing punishments.
With the vigorous development of intelligent campus construction, great changes have taken place in the development of information technology in colleges and universities from the previous digital to intelligent devel...
详细信息
Banknote counterfeiting is a common practice worldwide. Due to the recent developments in technology, banknote imitation has become easier than before. There are different kinds of algorithms developed for the detecti...
详细信息
Regression algorithm used for the prediction of output with given features and it is a supervised learning algorithm. In applying regression algorithms such as linear regression, Regression using ANN, Regression using...
详细信息
We derive the rate of convergence to the globally strongly variationally stable Nash equilibrium in a convex game, for a zeroth-order learning algorithm. Though we do not assume strong monotonicity of the game, our ra...
详细信息
ISBN:
(纸本)9798331540920;9783907144107
We derive the rate of convergence to the globally strongly variationally stable Nash equilibrium in a convex game, for a zeroth-order learning algorithm. Though we do not assume strong monotonicity of the game, our rates for the one-point feedback, O (Ndt(1/2)), and for the two-point feedback, O(N(2)d(2)t), match the best known rates for strongly monotone games under zeroth-order information.
Decentralized learning has emerged as an alternative method to the popular parameter-server framework which suffers from high communication burden, single-point failure and scalability issues due to the need of a cent...
详细信息
ISBN:
(纸本)1577358872
Decentralized learning has emerged as an alternative method to the popular parameter-server framework which suffers from high communication burden, single-point failure and scalability issues due to the need of a central server. However, most existing works focus on a single shared model for all workers regardless of the data heterogeneity problem, rendering the resulting model performing poorly on individual workers. In this work, we propose a novel personalized decentralized learning algorithm named DePRL via shared representations. Our algorithm relies on ideas from representation learning theory to learn a low-dimensional global representation collaboratively among all workers in a fully decentralized manner, and a user-specific low-dimensional local head leading to a personalized solution for each worker. We show that DePRL achieves, for the first time, a provable linear speedup for convergence with general non-linear representations (i.e., the convergence rate is improved linearly with respect to the number of workers). Experimental results support our theoretical findings showing the superiority of our method in data heterogeneous environments.
Recent success in training artificial agents and robots derives from a combination of direct learning of behavioural policies and indirect learning through value functions(1-3). Policy learning and value learning use ...
详细信息
Recent success in training artificial agents and robots derives from a combination of direct learning of behavioural policies and indirect learning through value functions(1-3). Policy learning and value learning use distinct algorithms that optimize behavioural performance and reward prediction, respectively. In animals, behavioural learning and the role of mesolimbic dopamine signalling have been extensively evaluated with respect to reward prediction(4);however, so far there has been little consideration of how direct policy learning might inform our understanding(5). Here we used a comprehensive dataset of orofacial and body movements to understand how behavioural policies evolved as naive, head-restrained mice learned a trace conditioning paradigm. Individual differences in initial dopaminergic reward responses correlated with the emergence of learned behavioural policy, but not the emergence of putative value encoding for a predictive cue. Likewise, physiologically calibrated manipulations of mesolimbic dopamine produced several effects inconsistent with value learning but predicted by a neural-network-based model that used dopamine signals to set an adaptive rate, not an error signal, for behavioural policy learning. This work provides strong evidence that phasic dopamine activity can regulate direct learning of behavioural policies, expanding the explanatory power of reinforcement learning models for animal learning(6). Analysis of data collected from mice learning a trace conditioning paradigm shows that phasic dopamine activity in the brain can regulate direct learning of behavioural policies, and dopamine sets an adaptive learning rate rather than an error-like teaching signal.
Capsule networks (see Hinton et al., 2018) aim to encode knowledge of and reason about the relationship between an object and its parts. In this letter, we specify a generative model for such data and derive a variati...
详细信息
Capsule networks (see Hinton et al., 2018) aim to encode knowledge of and reason about the relationship between an object and its parts. In this letter, we specify a generative model for such data and derive a variational algorithm for inferring the transformation of each model object in a scene and the assignments of observed parts to the objects. We derive a learning algorithm for the object models, based on variational expectation maximization (Jordan et al., 1999). We also study an alternative inference algorithm based on the RANSAC method of Fischler and Bolles (1981). We apply these inference methods to data generated from multiple geometric objects like squares and triangles ("constellations") and data from a parts-based model of faces. Recent work by Kosiorek et al. (2019) has used amortized inference via stacked capsule autoencoders to tackle this problem;our results show that we significantly outperform them where we can make comparisons (on the constellations data).
暂无评论