Knowledge extraction and retention based continual learning by using convolutional autoencoder-based learning classifier system

A New Ensemble Learning Method for Multiple Fusion Weighted Evidential Reasoning Rule

Ensemble learning, as a kind of method to improve the generalization ability of classifiers, is often used to improve the model effect in the field of deep learning. However, the present ensemble learning methods mostly adopt voting fusion in combining strategies. This strategy has difficulty mining effective information from the classifiers and cannot effectively reflect the relationship between different classifiers. Ensemble learning based on the evidential inference rule (ER rule) can effectively excavate the internal relationships among different classifiers and has a certain interpretability. However, the ER rule depends on the weight distribution of different combination strategies, and the setting of the evidence weight will affect the accuracy and stability of the model. Therefore, this paper proposes a new ensemble learning method based on multiple fusion weighted evidential reasoning rules and constructs an ensemble learning framework for data fusion and decision mapping. This framework takes the evidence weight, confidence, and feature data of each classifier as input and the integration results as output. The weight of evidence was determined by multiple fusion weights of the entropy weight method and order relation method. Finally, the integrated learning process is set up by the ER algorithm. The method proposed in this paper is verified by multiple datasets. Experimental results show that the surface construction model has good performance, and the defects of single weighting instability are greatly improved under the premise of improving the integration effect.

Online continual learning with declarative memory

Deep neural networks are enjoying unprecedented attention and success in recent years. However, catastrophic forgetting undermines the performance of deep models when the training data are arrived sequentially in an online multi-task learning fashion. To address this issue, we propose a novel method named continual learning with declarative memory (CLDM) in this paper. Specifically, our idea is inspired by the structure of human memory. Declarative memory is a major component of long-term memory which helps human beings memorize past experiences and facts. In this paper, we propose to formulate declarative memory as task memory and instance memory in neural networks to overcome catastrophic forgetting. Intuitively, the instance memory recalls the input-output relations (fact) in previous tasks, which is implemented by jointly rehearsing previous samples and learning current tasks as replaying-based methods act. In addition, the task memory aims to capture long-term task correlation information across task sequences to regularize the learning of the current task, thus preserving task-specific weight realizations (experience) in high task-specific layers. In this work, we implement a concrete instantiation of the proposed task memory by leveraging a recurrent unit. Extensive experiments on seven continual learning benchmarks verify that our proposed method is able to outperform previous approaches with tremendous improvements by retaining the information of both samples and tasks.

A Comprehensive "Real-World Constraints"-Aware Requirements Engineering Related Assessment and a Critical State-of-the-Art Review of the Monitoring of Humans in Bed

Currently, abnormality detection and/or prediction is a very hot topic. In this paper, we addressed it in the frame of activity monitoring of a human in bed. This paper presents a comprehensive formulation of a requirements engineering dossier for a monitoring system of a "human in bed" for abnormal behavior detection and forecasting. Hereby, practical and real-world constraints and concerns were identified and taken into consideration in the requirements dossier. A comprehensive and holistic discussion of the anomaly concept was extensively conducted and contributed to laying the ground for a realistic specifications book of the anomaly detection system. Some systems engineering relevant issues were also briefly addressed, e.g., verification and validation. A structured critical review of the relevant literature led to identifying four major approaches of interest. These four approaches were evaluated from the perspective of the requirements dossier. It was thereby clearly demonstrated that the approach integrating graph networks and advanced deep-learning schemes (Graph-DL) is the one capable of fully fulfilling the challenging issues expressed in the real-world conditions aware specification book. Nevertheless, to meet immediate market needs, systems based on advanced statistical methods, after a series of adaptations, already ensure and satisfy the important requirements related to, e.g., low cost, solid data security and a fully embedded and self-sufficient implementation. To conclude, some recommendations regarding system architecture and overall systems engineering were formulated.