 This research has shown that LSTM neural networks can be used to accurately predict lower limb joint torques during daily activities. This is achieved by first training the network on muscle electromyography signals and lower limb joint angles, then testing it on different subjects and or tasks. By transferring the learned knowledge from one subject slash task to another, the accuracy of prediction can be improved even further. This article was authored by Longbin Zhang, David Sosilia, Ruoli Wang, and others.