This article aims to explore the application and effect of Deep Learning (DL) in tertiary education quality assessment to build an objective, efficient, and accurate education quality assessment system. Aiming at the problems of traditional assessment methods, such as long period, high cost, and easily influenced by human factors, this article puts forward a DL-based tertiary education quality assessment model. By collecting the educational data (including academic performance, attendance records, homework completion, online learning behavior, etc.) of different professional courses in many universities, it carries out data preprocessing and feature engineering; designs a Convolutional Neural Network structure, including input layer, hidden layer, and output layer; and trains and optimizes the model. Experimental verification shows that the DL-based tertiary education quality assessment model is superior to the decision tree assessment method in key assessment indexes such as accuracy, recall, and F1 score. Additionally, the assessment speed is highly improved. This result fully proves the superiority and practicability of DL technology in tertiary education quality assessment and provides timely and accurate data support for educational decision making.