<![CDATA[This study aims to analyze the impact of deep learning implementation in STEM-based mathematics learning and identify challenges faced in its implementation. Referring to various recent studies, the deep learning approach has been proven to significantly improve conceptual understanding, critical thinking skills, mathematical problem-solving skills, and student motivation and engagement. The integration of deep learning with STEM enables contextual, collaborative, and real-world problem-oriented learning, which aligns with the principles of Outcome-Based Education (OBE) and the demands of 21st-century competencies. The method used in this study is a systematic literature review of several leading national and international journals published between 2019 and 2025. The analysis instruments include thematic identification of empirical findings, learning strategies, and supporting and inhibiting factors for implementation. The synthesis results show that despite its effectiveness, the implementation of deep learning in STEM-based mathematics learning faces several challenges, including limited technological infrastructure, low teacher readiness and digital literacy, dual cognitive load on students, time constraints in the curriculum, and ethical issues related to data privacy when involving artificial intelligence (AI). Thus, an integrative strategy is needed that includes teacher training, provision of supporting resources, and policies that support technology-based adaptive learning in an ethical and sustainable manner.]]>
Copyrights © 2026
