• Publications
  • Influence
Graph Theoretic Methods in Multiagent Networks
Global Convergence of Policy Gradient Methods for the Linear Quadratic Regulator
TLDR
This work bridges the gap showing that (model free) policy gradient methods globally converge to the optimal solution and are efficient (polynomially so in relevant problem dependent quantities) with regards to their sample and computational complexities.
Controllability of Multi-Agent Systems from a Graph-Theoretic Perspective
TLDR
This work shows how the symmetry structure of the network, characterized in terms of its automorphism group, directly relates to the controllability of the corresponding multi-agent system.
On maximizing the second smallest eigenvalue of a state-dependent graph Laplacian
TLDR
The motivation in the present work is to "assign" this Laplacian eigenvalue when relative positions of various elements dictate the interconnection of the underlying weighted graph, so as to "synthesize" information graphs that have desirable system theoretic properties.
Agreement over random networks
  • Yuko Hatano, M. Mesbahi
  • Computer Science, Mathematics
    43rd IEEE Conference on Decision and Control (CDC…
  • 2004
TLDR
In a random network, the existence of an information channel between a pair of elements at each time instance is probabilistic and independent of other channels; hence, the topology of the network varies over time.
Edge Agreement: Graph-Theoretic Performance Bounds and Passivity Analysis
TLDR
The dynamics induced by the edge Laplacian facilitates a better understanding of the role of certain subgraphs in the original agreement problem, and is employed to provide new insights into the nonlinear extension of linear agreement to agents with passive dynamics.
From Noisy Data to Feedback Controllers: Nonconservative Design via a Matrix S-Lemma
TLDR
A new method to obtain feedback controllers of an unknown dynamical system directly from noisy input/state data and derive nonconservative design methods for quadratic stabilization, and data-based linear matrix inequalities, enables control design from large datasets.
Online distributed optimization via dual averaging
TLDR
A distributed algorithm based on dual subgradient averaging is extended and yields an upper bound on regret as a function of the underlying network topology, specifically its connectivity.
On strong structural controllability of networked systems: A constrained matching approach
TLDR
An O(n2) algorithm to validate if a set of inputs leads to a strongly structurally controllable network and to find such an input set and the problem of finding such a set with minimal cardinality is shown to be NP-complete.
Global Convergence of Policy Gradient Methods for Linearized Control Problems
TLDR
This work bridges the gap showing that (model free) policy gradient methods globally converge to the optimal solution and are efficient (polynomially so in relevant problem dependent quantities) with regards to their sample and computational complexities.
...
...