Optimal Attack against Autoregressive Models by Manipulating the Environment
This addresses security vulnerabilities in time series forecasting models, but it is incremental as it applies known control methods to a specific threat model.
The paper tackles the problem of adversarial attacks on autoregressive time series forecasts by formulating an optimal attack using Linear Quadratic Regulator (LQR) and Model Predictive Control (MPC), with experiments showing effectiveness.
We describe an optimal adversarial attack formulation against autoregressive time series forecast using Linear Quadratic Regulator (LQR). In this threat model, the environment evolves according to a dynamical system; an autoregressive model observes the current environment state and predicts its future values; an attacker has the ability to modify the environment state in order to manipulate future autoregressive forecasts. The attacker's goal is to force autoregressive forecasts into tracking a target trajectory while minimizing its attack expenditure. In the white-box setting where the attacker knows the environment and forecast models, we present the optimal attack using LQR for linear models, and Model Predictive Control (MPC) for nonlinear models. In the black-box setting, we combine system identification and MPC. Experiments demonstrate the effectiveness of our attacks.