Wei Jiang

Wei Jiang, Gang Li, Yibo Wang et al.

Variance reduction techniques such as SPIDER/SARAH/STORM have been extensively studied to improve the convergence rates of stochastic non-convex optimization, which usually maintain and update a sequence of estimators for a single function across iterations. What if we need to track multiple functional mappings across iterations but only with access to stochastic samples of $\mathcal{O}(1)$ functional mappings at each iteration? There is an important application in solving an emerging family of coupled compositional optimization problems in the form of $\sum_{i=1}^m f_i(g_i(\mathbf{w}))$, where $g_i$ is accessible through a stochastic oracle. The key issue is to track and estimate a sequence of $\mathbf g(\mathbf{w})=(g_1(\mathbf{w}), \ldots, g_m(\mathbf{w}))$ across iterations, where $\mathbf g(\mathbf{w})$ has $m$ blocks and it is only allowed to probe $\mathcal{O}(1)$ blocks to attain their stochastic values and Jacobians. To improve the complexity for solving these problems, we propose a novel stochastic method named Multi-block-Single-probe Variance Reduced (MSVR) estimator to track the sequence of $\mathbf g(\mathbf{w})$. It is inspired by STORM but introduces a customized error correction term to alleviate the noise not only in stochastic samples for the selected blocks but also in those blocks that are not sampled. With the help of the MSVR estimator, we develop several algorithms for solving the aforementioned compositional problems with improved complexities across a spectrum of settings with non-convex/convex/strongly convex/Polyak-Łojasiewicz (PL) objectives. Our results improve upon prior ones in several aspects, including the order of sample complexities and dependence on the strong convexity parameter. Empirical studies on multi-task deep AUC maximization demonstrate the better performance of using the new estimator.

Wei Jiang, Zhaoxia Peng, Ahmed Rahmani et al.

The problem of disturbance rejection/attenuation for constant-input delayed linear multi-agent systems (MASs) with the directed communication topology is tackled in this paper, where a classic model reduction technique is introduced to transform the delayed MAS into the delay-free one. First, when the leader has no control input, a novel adaptive predictive extended state observer (ESO) using only relative state information of neighboring agents is designed to achieve disturbance-rejected consensus tracking. The stabilization analysis is presented via the Lyapunov function and sufficient conditions are derived in terms of linear matrix inequalities. Then the result is extended to the disturbance-attenuated case where the leader has bounded control input which is only known by a portion of followers. Finally, two numerical examples are presented to illustrate the effectiveness of proposed strategies. The main contribution focuses on the design of adaptive predictive ESO protocols with the fully distributed property.

Mohammadreza Doostmohammadian, Wei Jiang, Muwahida Liaquat et al.

We consider a discrete-time model of continuous-time distributed optimization over dynamic directed-graphs (digraphs) with applications to distributed learning. Our optimization algorithm works over general strongly connected dynamic networks under switching topologies, e.g., in mobile multi-agent systems and volatile networks due to link failures. Compared to many existing lines of work, there is no need for bi-stochastic weight designs on the links. The existing literature mostly needs the link weights to be stochastic using specific weight-design algorithms needed both at the initialization and at all times when the topology of the network changes. This paper eliminates the need for such algorithms and paves the way for distributed optimization over time-varying digraphs. We derive the bound on the gradient-tracking step-size and discrete time-step for convergence and prove dynamic stability using arguments from consensus algorithms, matrix perturbation theory, and Lyapunov theory. This work, particularly, is an improvement over existing stochastic-weight undirected networks in case of link removal or packet drops. This is because the existing literature may need to rerun time-consuming and computationally complex algorithms for stochastic design, while the proposed strategy works as long as the underlying network is weight-symmetric and balanced. The proposed optimization framework finds applications to distributed classification and learning.

Wei Jiang, Zhaoxia Peng, Guoguang Wen et al.

This paper presents a unified framework of time-varying formation (TVF) design for general linear multi-agent systems (MAS) based on an observer viewpoint from undirected to directed topology, from stabilization to tracking and from a leader without input to a one with bounded input. The followers can form a TVF shape which is specified by piecewise continuously differential vectors. The leader's trajectory, which is available to only a subset of followers, is also time-varying. For the undirected formation tracking and directed formation stabilization cases, only the relative output measurements of neighbors are required to design control protocols; for the directed formation tracking case, the agents need to be introspective (i.e. agents have partial knowledge of their own states) and the output measurements are required. Furthermore, considering the real applications, the leader with bounded input case is studied. One main contribution of this paper is that fully distributed adaptive output protocols, which require no global information of communication topology and do not need the absolute or relative state information, are proposed to solve the TVF control problem. Numerical simulations including an application to nonholonomic mobile vehicles are provided to verify the theoretical results.

Wei Jiang, Hans Dieter Schotten

Opportunistic relay selection (ORS) has been recognized as a simple but efficient method for mobile nodes to achieve cooperative diversity in slow fading channels. However, the wrong selection of the best relay arising from outdated channel state information (CSI) in fast time-varying channels substantially degrades its performance. With the proliferation of high-mobility applications and the adoption of higher frequency bands in 5G and beyond systems, the problem of outdated CSI will become more serious. Therefore, the design of a novel cooperative method that is applicable to not only slow fading but also fast fading is increasingly of importance. To this end, we develop and analyze a deep-learning-aided cooperative method coined predictive relay selection (PRS) in this article. It can remarkably improve the quality of CSI through fading channel prediction while retaining the simplicity of ORS by selecting a single opportunistic relay so as to avoid the complexity of multi-relay coordination and synchronization. Information-theoretic analysis and numerical results in terms of outage probability and channel capacity reveal that PRS achieves full diversity gain in slow fading wireless environments and substantially outperforms the existing schemes in fast fading channels.

Yan Wei, Wei Jiang, Ahmed Rahmani et al.

A high redundant non-holonomic humanoid mobile dual-arm manipulator system is presented in this paper where the motion planning to realize "human-like" autonomous navigation and manipulation tasks is studied. Firstly, an improved MaxiMin NSGA-II algorithm, which optimizes five objective functions to solve the problems of singularity, redundancy, and coupling between mobile base and manipulator simultaneously, is proposed to design the optimal pose to manipulate the target object. Then, in order to link the initial pose and that optimal pose, an off-line motion planning algorithm is designed. In detail, an efficient direct-connect bidirectional RRT and gradient descent algorithm is proposed to reduce the sampled nodes largely, and a geometric optimization method is proposed for path pruning. Besides, head forward behaviors are realized by calculating the reasonable orientations and assigning them to the mobile base to improve the quality of human-robot interaction. Thirdly, the extension to on-line planning is done by introducing real-time sensing, collision-test and control cycles to update robotic motion in dynamic environments. Fourthly, an EEs' via-point-based multi-objective genetic algorithm is proposed to design the "human-like" via-poses by optimizing four objective functions. Finally, numerous simulations are presented to validate the effectiveness of proposed algorithms.