||Despite the remarkable success in a broad set of sensing appli- cations, state-of-the-art deep learning techniques struggle with complex reasoning tasks across a distributed set of sensors. Unlike recognizing transient complex activities (e.g., human activities such as walking or running) from a single sensor, detecting more complex events with larger spatial and temporal dependencies across multi- ple sensors is extremely difficult, e.g., utilizing a hospital's sensor network to detect whether a nurse is following a sanitary protocol as they traverse from patient to patient. Training a more compli- cated model requires a larger amount of data–which is unrealistic considering complex events rarely happen in nature. Moreover, neu- ral networks struggle with reasoning about serial, aperiodic events separated by large quantities in the spatial-temporal dimensions. We propose Neuroplex, a neural-symbolic framework that learns to perform complex reasoning on raw sensory data with the help of high-level, injected human knowledge. Neuroplex decomposes the entire complex learning space into explicit perception and rea- soning layers, i.e., by maintaining neural networks to perform low- level perception tasks and neurally reconstructed reasoning models to perform high-level, explainable reasoning. After training the neurally reconstructed reasoning model using human knowledge, Neuroplex allows effective end-to-end training of perception mod- els with an additional semantic loss using only sparse, high-level annotations. Our experiments and evaluation show that Neuroplex is capable of learning to efficiently and effectively detect complex events–which cannot be handled by state-of-the-art neural network models. During the training, Neuroplex not only reduces data an- notation requirements by 100x, but also significantly speeds up the learning process for complex event detection by 4x.