Computation of non-linear functions such as hyperbolic tangent and sigmoid is critical in machine learning. Furthermore, precision requirement varies across different layers in deep neural networks.

Economic productivity is shown to peak at an annual average temperature of 13 °C and decline at high temperatures, indicating that climate change is expected to lower global incomes more than 20% ...

In order to make the LuT learning sample-adaptive, we parameterize the intensity mapping function by exploring two families of non-linear functions (i.e., piece-wise linear and power functions) and ...

We study reward-free reinforcement learning (RL) under general non-linear function approximation, and establish sample efficiency and hardness results under various standard structural assumptions. On ...