Continual Learning with Gated Incremental Memories for sequential data processing

By Andrea Cossu, Antonio Carta, Davide Bacciu. 2020

27 citations recorded by the source corpus.

The ability to learn in dynamic, nonstationary environments without forgetting previous knowledge, also known as Continual Learning (CL), is a key enabler for scalable and trustworthy deployments of adaptive solutions. While the importance of continual learning is largely acknowledged in machine vision and reinforcement learning problems, this is mostly under-documented for sequence processing tasks. This work proposes a Recurrent Neural Network (RNN) model for CL that is able to deal with concept drift in input distribution without forgetting previously acquired knowledge. We also implement and test a popular CL approach, Elastic Weight Consolidation (EWC), on top of two different types of RNNs. Finally, we compare the performances of our enhanced architecture against EWC and RNNs on a set of standard CL benchmarks, adapted to the sequential data processing scenario. Results show the superior performance of our architecture and highlight the need for special solutions designed to address CL in RNNs.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Electrical and Electronic Engineering

Citations in the corpus

GCR: Gradient Coreset based Replay Buffer Selection for Continual Learning (2021, neutral): “Architecture adjustments modify model architectures to incorporate CL challenges; for instance, using a recurrent neural network [13], learning overcapacitated deep neural networks in an adaptive way that repeatedly compresses sparse connections [43] or using overlapping convolutional filters [18].”
Organizing recurrent network dynamics by task-computation to enable continual learning (2020, background): “Recently, approaches to continual learning based on dynamical network architectures have also been developed in the RNN setting [23, 24].”
Continual Learning in Recurrent Neural Networks with Hypernetworks (2020, background): “Another alternative, utilized in [20, 50] on sequential data, is the use of a different autoencoder per task.”
Continual learning in recurrent neural networks (2020, background): “• We identify elements that critically affect the stability-plasticity dilemma of weight-importance methods in RNNs.”