<![CDATA[Mathematical problem-solving ability is one of the essential competencies that students need to master in mathematics learning. However, many students still experience difficulties in understanding mathematical concepts and applying appropriate strategies to solve problems. This study aimed to determine the effect of Self-Directed Learning in Deep Learning-based Problem Based Learning on students' mathematical problem-solving abilities. This research employed a quantitative approach with a causal associative design. The sample consisted of 36 students of class X-5 at SMAN 1 Muncar selected through purposive sampling. Data were collected using Self-Directed Learning questionnaires, Deep Learning-based Problem Based Learning questionnaires, and mathematical problem-solving tests. The data were analyzed using descriptive statistics and multiple linear regression analysis with the assistance of IBM SPSS Statistics 26. The results showed that Self-Directed Learning significantly affected students' mathematical problem-solving abilities with a significance value of 0.005 < 0.05. Deep Learning-based Problem Based Learning also significantly affected students' mathematical problem-solving abilities with a significance value of 0.032 < 0.05. Simultaneously, both variables significantly influenced students' mathematical problem-solving abilities with a significance value of 0.000 < 0.05 and an F value of 966.260. Therefore, Self-Directed Learning and Deep Learning-based Problem Based Learning can be used as effective learning strategies to improve students' mathematical problem-solving abilities.]]>
