Arash Bahari Kordabad

Dr. Arash Bahari Kordabad


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 November 2025 in arXiv
Verified
Preprint
Open access
PDF file
AKArash Bahari Kordabad
Arash Bahari Kordabad
DBDean Brandner
Dean Brandner
SGSebastien Gros
Sebastien Gros
SLSergio Lucia
Sergio Lucia
SSSadegh Soudjani
Sadegh Soudjani
In this paper, we propose a second-order deterministic actor-critic framework in reinforcement learning that extends the classical deterministic policy gradient method to exploit curvature information of the performance function. Building on the concept of compatible function approximation for the critic, we introduce a quadratic critic that simultaneously preserves the true policy gradient and an approximation of the performance Hessian. A least-squares temporal difference learning scheme is then developed to estimate the quadratic critic parameters efficiently. This construction enables a quasi-Newton actor update using information learned by the critic, yielding faster convergence compared to first-order methods. The proposed approach is general and applicable to any differentiable policy class. Numerical examples demonstrate that the method achieves improved convergence and performance over standard deterministic actor-critic baselines.

Verified
Journal Article
Open access
PDF file
Hossein Nejatbakhsh EsfahaniHossein Nejatbakhsh Esfahani
Hossein Nejatbakhsh Esfahani
AKArash Bahari Kordabad
Arash Bahari Kordabad
David Moreno-SalinasDavid Moreno-Salinas
David Moreno-Salinas
Autonomous and highly automated maritime systems are moving from controlled demonstrations to sustained operations in congested waterways and harsh sea states

3 November 2025 in arXiv
Verified
Preprint
Open access
PDF file
AVAmirreza Valaei
Amirreza Valaei
AKArash Bahari Kordabad
Arash Bahari Kordabad
SSSadegh Soudjani
Sadegh Soudjani
Policy gradient methods are a powerful family of reinforcement learning algorithms for continuous control that optimize a policy directly. However, standard first-order methods often converge slowly. Second-order methods can accelerate learning by using curvature information, but they are typically expensive to compute. The linear quadratic regulator (LQR) is a practical setting in which key quantities, such as the policy gradient, admit closed-form expressions. In this work, we develop second-order policy gradient algorithms for LQR by deriving explicit formulas for both the approximate and exact Hessians used in Gauss--Newton and Newton methods, respectively. Numerical experiments show a faster convergence rate for the proposed second-order approach over the standard first-order policy gradient baseline.

29 October 2025 in arXiv
Verified
Preprint
Open access
PDF file
AKArash Bahari Kordabad
Arash Bahari Kordabad
RMRupak Majumdar
Rupak Majumdar
SSSadegh Soudjani
Sadegh Soudjani
In this paper, we present a computational approach to certify almost sure reachability for discrete-time polynomial stochastic systems by turning drift--variant criteria into sum-of-squares (SOS) programs solved with standard semidefinite solvers. Specifically, we provide an SOS method based on two complementary certificates: (i) a drift certificate that enforces a radially unbounded function to be non-increasing in expectation outside a compact set of states; and (ii) a variant certificate that guarantees a one-step decrease with positive probability and ensures the target contains its nonpositive sublevel set. We transform these conditions to SOS constraints. For the variant condition, we enforce a robust decrease over a parameterized disturbance ball with nonzero probability and encode the constraints via an S-procedure with polynomial multipliers. The resulting bilinearities are handled by an alternating scheme that alternates between optimizing multipliers and updating the variant and radius until a positive slack is obtained. Two case studies illustrate the workflow and certifies almost-sure reachability.

19 February 2026 in arXiv
Verified
Preprint
Open access
PDF file
EVEleftherios E. Vlahakis
Eleftherios E. Vlahakis
AKArash Bahari Kordabad
Arash Bahari Kordabad
LLLars Lindemann
Lars Lindemann
PSPantelis Sopasakis
Pantelis Sopasakis
SSSadegh Soudjani
Sadegh Soudjani
DDDimos V. Dimarogonas
Dimos V. Dimarogonas
Multi-agent planning under Signal Temporal Logic (STL) is often hindered by collaborative tasks that lead to computational challenges due to the inherent high-dimensionality of the problem, preventing scalable synthesis with satisfaction guarantees. To address this, we formulate STL planning as an optimization program under arbitrary multi-agent constraints and introduce a penalty-based unconstrained relaxation that can be efficiently solved via a Block-Coordinate Gradient Descent (BCGD) method, where each block corresponds to a single agent's decision variables, thereby mitigating complexity. By utilizing a quadratic penalty function defined via smooth STL semantics, we show that BCGD iterations converge to a stationary point of the penalized problem under standard regularity assumptions. To enforce feasibility, the BCGD solver is embedded within a two-layer optimization scheme: inner BCGD updates are performed for a fixed penalty parameter, which is then increased in an outer loop to progressively improve multi-agent STL robustness. The proposed framework enables scalable computations and is validated through various complex multi-robot planning scenarios.

24 December 2024 in arXiv
Verified
Preprint
Open access
PDF file
AAAkhil S Anand
Akhil S Anand
AKArash Bahari Kordabad
Arash Bahari Kordabad
Mario ZanonMario Zanon
Mario Zanon
SGSebastien Gros
Sebastien Gros
Optimality is a critical aspect of Model Predictive Control (MPC), especially in economic MPC. However, achieving optimality in MPC presents significant challenges, and may even be impossible, due to inherent inaccuracies in the predictive models. Predictive models often fail to accurately capture the true system dynamics, such as in the presence of stochasticity, leading to suboptimal MPC policies. In this paper, we establish the necessary and sufficient conditions on the underlying prediction model for an MPC scheme to achieve closed-loop optimality. Interestingly, these conditions are counterintuitive to the traditional approach of building predictive models that best fit the data. These conditions present a mathematical foundation for constructing models that are directly linked to the performance of the resulting MPC scheme.