Continuous learning method of sequence recommendation model based on sample playback

Abstract

The invention relates to a continuous learning method of a sequence recommendation model based on sample playback, which relates to the technical field of sequence recommendation, and comprises the following steps: 1, constructing a sequence recommendation model, and training the sequence recommendation model by using initial data; 2, sampling a small part of representative example samples based on a sample selection strategy of article category balance; 3, calculating and storing soft labels of the sampled example samples so as to participate in calculation of a distillation loss function part in next model updating; 4, providing an accurate recommendation service for a user by using the sequence recommendation model, and collecting new data obtained in a new period at the same time; and 5, updating parameters of the sequence recommendation model by using new data obtained in a new period and previously stored sampled examples The problem of disastrous forgetting faced by using a neural network sequence recommendation model in a continuous learning scene is effectively solved.

Description

technical field [0001] The invention relates to the technical field of sequence recommendation, in particular to a continuous learning method of a sequence recommendation model based on sample playback. Background technique [0002] In recent years, research on the algorithm design and practical application of sequential recommendation systems has attracted widespread attention in both academia and industry. With the introduction and use of deep learning technology, sequential recommendation algorithms based on deep learning have more powerful feature mining capabilities than traditional recommendation algorithms such as collaborative filtering and factorization machines, so they can effectively capture and utilize user interest preferences The trend of change, and then generate better recommendation results. [0003] Although existing neural network sequence recommendation models have achieved sufficient success, they are usually trained and tested offline using static dat...

AI Technical Summary

Problems solved by technology

This method makes it very easy for the model to forget the previously learned user preference patterns, leading to the occurrence of catastrophic forgetting problems

This will greatly reduce the recommendation effect of the recommendation system deployed in the continuous learning scenario

Images