Important concerns when using medical data for Machine Learning (ML) is patient privacy and bias. Federated Learning (FL), the training of a centralized model by using parameters from decentralized models, is alternatively used to protect patient privacy. Medical data can often be structured and sparse, where deep learning is not applicable. In this work, we applied Federated Learning with Logistic Regression on medical data through multiple experiments with data distribution among clients. Three simulated cluster sampling methods conducted to compare model accuracy with different data distributions and/or sample sizes. Our observations were as follows: (1) the Federated Learning model performs, on average, better than the average of individual clients, and (2) variability increases as the sample size and the number of clients increases, and accuracy stays moderately consistent. In addition, we designed a Federated Learning system that securely transfers model parameters between server and clients. We also present an approach to study bias in these models. Our results show that Federated Learning is a promising approach for medical applications.

In this work, we propose the notion of a Parse Condition—a logical condition that is satisfiable if and only if a given string w can be successfully parsed using a grammar G. Further, we propose an algorithm for building an SMT encoding of such parse conditions for LL(1) grammars and demonstrate its utility by building two applications over it: automated repair of syntax errors in Tiger programs and automated parser synthesis to automatically synthesize LL(1) parsers from examples. We implement our ideas into a tool, Cyclops, that is able to successfully repair 80% of our benchmarks (675 buggy Tiger programs), clocking an average of 30 seconds per repair and synthesize parsers for interesting languages from examples. Like verification conditions (encoding a program in logic) have found widespread applications in program analysis, we believe that Parse Conditions can serve as a foundation for interesting applications in syntax analysis.