In this paper, a new Messenger Routing Problem (MRP) is studied, which is motivated by Unmanned Aerial Vehicles (UAVs) accessing Unmanned Ground Vehicles (UGVs) to deliver information in UAV-UGV coordination systems. The objective is to minimize the longest path among multiple messengers, ensuring fast and reliable information transmission. Two key challenges arise in tackling this problem. First, the targets are moving, incurring the travel cost between targets to vary with the travel process. Second, the messengers accessing the neighborhood of targets needs to satisfy the communication time constraint. Based on the idea of decoupling, a two-phase planner is proposed to sequentially determine global access sequence and optimize local access locations. In the first phase, a motion prediction module is introduced in the Adaptive Large Neighborhood Search (ALNS) framework to deal with the dynamic characteristics of MRP. In the second phase, an efficient bisection sampling method based on the prediction points is proposed to obtain a shorter access path while satisfying the communication time constraint. Finally, the effectiveness and efficiency of the proposed method are demonstrated by performance evaluation and comparison with the state-of-the-art algorithms.

Evolutionary computing (EC), as a metaheuristic algorithm, is widely applied in optimization problems due to its effectiveness. During the optimization process of EC, a large amount of data will be generated, which contains rich information on the evolutionary process. Mining and utilizing these data can learn promising evolutionary knowledge to assist the EC algorithms in achieving more efficient evolution. First, we analyze in detail the key differences in collecting experience and guiding evolution between particle swarm optimization (PSO) and differential evolution (DE) in the existing knowledge learning framework. Then, we propose a triple network knowledge learning (TNKL) framework to mitigate the limitations and boost knowledge learning for PSO-based algorithms. In the TNKL framework, the successful evolution experience of the current particle, the historical best position, and the global best position are collected. Three neural networks learn different knowledge from these experiences and give appropriate evolution directions. The TNKL framework then comprehensively considers these directions according to the evolution stage to guide particle evolution. Finally, we integrate the TNKL framework with PSO and its state-of-the-art variants. The benchmark function experimental results verify the effectiveness and efficiency of the TNKL framework.

Distribution forecast can quantify forecast uncertainty and provide various forecast scenarios with their corresponding estimated probabilities. Accurate distribution forecast is crucial for planning – for example when making production capacity or inventory allocation decisions. We propose a practical and robust distribution forecast framework that relies on backtest-based bootstrap and adaptive residual selection. The proposed approach is robust to the choice of the underlying forecasting model, accounts for uncertainty around the input covariates, and relaxes the independence between residuals and covariates assumption. It reduces the Absolute Coverage Error by more than 63% compared to the classic bootstrap approaches and by 2%−32% compared to a variety of State-of-the-Art deep learning approaches on in-house product sales data and M4-hourly competition data.

Aiming at convex optimization under structural constraints, this work introduces and analyzes a variant of the Frank Wolfe (FW) algorithm termed ExtraFW. The distinct feature of ExtraFW is the pair of gradients leveraged per iteration, thanks to which the decision variable is updated in a prediction-correction (PC) format. Relying on no problem dependent parameters in the step sizes, the convergence rate of ExtraFW for general convex problems is shown to be ${\cal O}(\frac{1}{k})$, which is optimal in the sense of matching the lower bound on the number of solved FW subproblems. However, the merit of ExtraFW is its faster rate ${\cal O}\big(\frac{1}{k^2} \big)$ on a class of machine learning problems. Compared with other parameter-free FW variants that have faster rates on the same problems, ExtraFW has improved rates and fine-grained analysis thanks to its PC update. Numerical tests on binary classification with different sparsity-promoting constraints demonstrate that the empirical performance of ExtraFW is significantly better than FW, and even faster than Nesterov's accelerated gradient on certain datasets. For matrix completion, ExtraFW enjoys smaller optimality gap, and lower rank than FW.

This paper studies the adversarial graphical contextual bandits, a variant of adversarial multi-armed bandits that leverage two categories of the most common side information: contexts and side observations. In this setting, a learning agent repeatedly chooses from a set of K actions after being presented with a d-dimensional context vector. The agent not only incurs and observes the loss of the chosen action, but also observes the losses of its neighboring actions in the observation structures, which are encoded as a series of feedback graphs. This setting models a variety of applications in social networks, where both contexts and graph-structured side observations are available. Two efficient algorithms are developed based on EXP3. Under mild conditions, our analysis shows that for undirected feedback graphs the first algorithm, EXP3-LGC-U, achieves a sub-linear regret with respect to the time horizon and the average independence number of the feedback graphs. A slightly weaker result is presented for the directed graph setting as well. The second algorithm, EXP3-LGC-IX, is developed for a special class of problems, for which the regret is the same for both directed as well as undirected feedback graphs. Numerical tests corroborate the efficiency of proposed algorithms.

Cascading bandit (CB) is a popular model for web search and online advertising. However, the stationary CB model may be too simple to cope with real-world problems, where user preferences may change over time. Considering piecewise-stationary environments, two efficient algorithms, GLRT-CascadeUCB and GLRT-CascadeKL-UCB, are developed. Comparing with existing works, the proposed algorithms: i) are free of change-point-dependent information for choosing parameters; ii) have fewer tuning parameters; iii) improve regret upper bounds. We also show that the proposed algorithms are optimal up to logarithm terms by deriving a minimax lower bound $\Omega (\sqrt {NLT} )$ for piecewise-stationary CB. The efficiency of the proposed algorithms is validated through numerical tests on a real-world benchmark dataset.

We investigate the piecewise-stationary combinatorial semi-bandit problem. Compared to the original combinatorial semi-bandit problem, our setting assumes the reward distributions of base arms may change in a piecewise-stationary manner at unknown time steps. We propose an algorithm, GLR-CUCB, which incorporates an efficient combinatorial semi-bandit algorithm, CUCB, with an almost parameter-free change-point detector, the Generalized Likelihood Ratio Test (GLRT). Our analysis shows that the regret of GLR-CUCB is upper bounded by O(√NKT log T), where N is the number of piecewise-stationary segments, K is the number of base arms, and T is the number of time steps. As a complement, we also derive a nearly matching regret lower bound on the order of Ω(√NKT), for both piecewise-stationary multi-armed bandits and combinatorial semi-bandits, using information-theoretic techniques and judiciously constructed piecewise-stationary bandit instances. Our lower bound is tighter than the best available regret lower bound, which is Ω(√T). Numerical experiments on both synthetic and real-world datasets demonstrate the superiority of GLR-CUCB compared to other state-of-the-art algorithms.

The variance reduction class of algorithms including the representative ones, SVRG and SARAH, have well documented merits for empirical risk minimization problems. However, they require grid search to tune parameters (step size and the number of iterations per inner loop) for optimal performance. This work introduces ‘almost tune-free’ SVRG and SARAH schemes equipped with i) Barzilai-Borwein (BB) step sizes; ii) averaging; and, iii) the inner loop length adjusted to the BB step sizes. In particular, SVRG, SARAH, and their BB variants are first reexamined through an ‘estimate sequence’ lens to enable new averaging methods that tighten their convergence rates theoretically, and improve their performance empirically when the step size or the inner loop length is chosen large. Then a simple yet effective means to adjust the number of iterations per inner loop is developed to enhance the merits of the proposed averaging schemes and BB step sizes. Numerical tests corroborate the proposed methods.

Quantum wireless communications (QWC) is gaining much attention for its incomparable advantages of secure key transmission in free-space scenarios. In this paper, a novel iterative multi-user quantum key distribution (MQKD) non-orthogonal multiple-access (NOMA) system over Poisson atmospheric channels is proposed. Considering the atmospheric attenuation and quantum direct product operation, a MQKD parallel interference cancellation (PIC) approach is designed at the receiver. In particular, a quantum nonlinear Poisson shot noise limited sum-product interference cancellation algorithm is derived. Simulation results show that the proposed scheme is capable of mitigating the multi-user interference effectively, even considering the polarization mismatch. Quantitatively, for the 4-user scenario, it can achieve the classical Bit Error Rate (BER) of 3 x 10-6 at 2.5 km distance with polarization mismatch factor of 0.05. Moreover, the proposed NOMA-MQKD scheme can have obviously higher secure key rate than the conventional WDMA-MQKD approach.

Quantum wireless communications (QWC) is emerging as a promising candidate for future terrestrial information networks. In this paper, over noisy quantum atmospheric turbulence channels, a new non-orthogonal coherentstate massive-MIMO uplink scheme is proposed. At the receiver, a composite quantum iterative multi-stage measuring instrument and MIMO detector is developed. In particular, a non-additive Poisson shot noise limited quantum iterative massive-MIMO interference cancellation algorithm is derived. Simulation results show that the proposed scheme is capable of obtaining the substantial spatial diversity gain in severely noisy quantum turbulence channels, while effectively mitigating the quantum multi-user (MU) interference. Quantitatively, for the 8 × 64 MIMO scenario, it can achieve the BER of 10-6 at -179.3dBJ per bit over a strong Gamma-Gamma turbulence fading channel (α = 2.1, βc = 2.4). Furthermore, the proposed scheme can provide significant performance improvements over a recent reported photon-counting MIMO receiver [18].

We consider a problem that recovers a 2-D object and the underlying view angle distribution from its noisy projection tilt series taken at unknown view angles. Traditional approaches rely on the estimation of the view angles of the projections, which do not scale well with the sample size and are sensitive to noise. We introduce a new approach using the moment features to simultaneously recover the underlying object and the distribution of view angles. This problem is formulated as constrained nonlinear least squares in terms of the truncated Fourier-Bessel expansion coefficients of the object and is solved by a new alternating direction method of multipliers (ADMM)-based algorithm. Our numerical experiments show that the new approach outperforms the expectation maximization (EM)-based maximum marginalized likelihood estimation in efficiency and accuracy. Furthermore, the hybrid method that uses EM to refine ADMM solution achieves the best performance.

Non-orthogonal multiple access (NOMA) comprises a prospective technology in the 5th generation (5G) and future wireless communications. In this paper, due to intrinsic nonorthogonal properties of coherent state multi-user communications, a new iterative quantum spatial coupling NOMA (q-SC-NOMA) scheme is proposed. The simulation results confirm that, compared to the recent reported multi-stage displacement based receiver (MDR), the q-SC-NOMA scheme is capable of achieving obviously better performance, saving about 3dB mean signal power in 3-user scenario. Moreover, the q-SC-NOMA scheme can obtain the lower bit error rate (BER) than the non-spatial-coupled quantum NOMA (q-NOMA) approach with reduced computational complexity. In particular, the q-SC-NOMA also offers high robustness against many practical quantum domain imperfections, e.g. mode mismatch and dark counts, etc.

One of the most crucial challenges in large-scale quantum communication systems is how to simultaneously transmit quantum information among multiple users via a common channel. In this paper, a novel non-orthogonal iterative quantum multiuser communication system is proposed. Based on the beam splitter network (BSN), we combine the independent quantum-state signals of different users into a single transmission channel. Furthermore, using photon-number-resolving detectors (PNRDs) based quantum adaptive feedback positive operator valued measurement, a chip-interleaved quantum iterative parallel interference cancellation (Iter-PIC) multiuser detection is developed. From the simulation results, the proposed quantum multiuser communication system can easily achieve 10e-7 bit error rate (BER) performance at 42 photons per bit for 6 users transmitting simultaneously. Also, from the extrinsic information transfer trajectories (EXIT) analysis results, this quantum chip interleaved Iter-PIC system is verified to have robust and fast convergence property in tough conditions, e.g. extra low receiving signal energy and quantum mode mismatch. Moreover, we show that the throughput of proposed quantum Iter-PIC system is 66.7% higher than that of the existing quantum wavelength division multiple access (q-WDMA) communication systems.

A new strategy of iterative quantum receivers with photon-number-resolving detectors is proposed to discriminate coherent states. The iterative quantum receiver provides a bit error rate closer to the Helstrom limit when compared with a recently proposed non-iterative quantum receiver. Moreover, the proposed iterative receiver is robust against various imperfections, such as mode mismatch, non-ideal quantum efficiency and dark counts. The simulation results indicate that the iterative receiver is capable of achieving 2.16 dB improvement upon the standard quantum limit, leaving only 0.65 dB from the Helstrom limit. Besides, with Turbo encoder and decoder, the iterative receiver gives about 1.46 dB improvement over the conventional non-iterative quantum receiver and approaches the equivalent Helstrom limit.