Next-generation wireless communication systems are expected to operate at higher carrier frequencies to meet the data rate requirements necessary for emerging use cases such as smart cities, e-health, and digital twins for manufacturing. Higher carrier frequencies also enable new functionalities, such as integrated sensing and communications (ISAC). ISAC aims to integrate radar and communication capabilities in one joint system, which enables hardware sharing, energy savings, communication in high-frequency radar bands, and improved channel estimation via sensing-assisted communications, among other advantages. However, conventional ISAC approaches degrade in performance under model mismatch, i.e., if the underlying reality does not match the assumed mathematical models. In particular at high carrier frequencies, hardware impairments can severely affect the system performance and hardware design becomes very challenging. In this project, we study machine learning in the context of ISAC under hardware impairments. The goal is to propose machine learning approaches to compensate for hardware impairments in ISAC, and compare them with state-of-the-art model-based solutions.