The criticality of efficient traffic forecasting in Intelligent Transportation System (ITS) has garnered significant academic attention. This study addresses the prevalent issue of distribution shift in real-world datasets, which often degrades performance, and explores the effectiveness of the channel-independence (CI), a technique recently proposed to mitigate this issue. While Spatio-Temporal Graph Neural Networks (STGNNs) are noted for their flexibility to represent road structures, their designs typically lack the capability to integrate CI without disrupting the spatial relationships, potentially limiting the performance. We present a novel approach that successfully integrates CI into spatial-temporal forecasting by incorporating distinct temporal, spatial, and predefined graph structure information within each channel. Moreover, STGNNs frequently emphasize intricate designs, which result in increased computational demands while offering only marginal improvements in accuracy. This paper presents ST-MLP, a streamlined spatio-temporal model constructed exclusively from cascaded Multi-Layer Perceptron (MLP) modules and linear layers. Experimental results indicate that ST-MLP outperforms numerous existing STGNNs in both accuracy and computational efficiency. Our findings advocate for further investigation into more streamlined and effective neural network architectures within spatial-temporal forecasting research.

We present MASRL: Competitive Multi-Agent Self-supervised representations for Reinforcement Learning in the multi-agent competitive environment. MASRL introduces a simple but effective self-supervised task: predicting a learning agent’s opponent’s future move. In doing this, the agent learns a stronger representation from this additional signal, focusing not only on itself but also on its opponent. By understanding and anticipating the opponent’s future moves, MASRL allows the learning agent to develop effective strategies for opponent exploitation. Our method stabilizes training, improves sample efficiency, and allows the agent to generalize and adapt its playing strategy to other unseen expert opponents. On the Multi-Agent Atari benchmark, MASRL achieves remarkable performance, outperforming other strong baselines. Examples of demo videos can be found at: https://sites.google.com/view/compmarl

Leading international financial firms are applying multi-stage stochastic programs for managing asset-liability risk over extended time periods. Prominent examples include: Towers Perrin, State Farm Insurance, Falcon Asset Management, Frank Russell and Unilever. The asset-liability management systems assist pension plan investors, banks, insurance companies and other leveraged institutions. Wealthy individuals can benefit by developing careful risk management strategies. The advantages of integrating assets and liabilities are discussed along with a brief comparison of alternative modeling frameworks. We describe the advantages of high-performance computers for solving these difficult nonlinear robust optimization problems.

This paper extends a primal-dual interior point procedure for linear programs to the case of convex separable quadratic objectives. Included are efficient procedures for: attaining primal and dual feasibility, variable upper bounding, and free variables. A sequential procedure that invokes the quadratic solver is proposed and implemented for solving linearly constrained convex separable nonlinear programs. Computational results are provided for several large test cases from stochastic programming. The proposed methods compare favorably with MINOS, especially for the larger examples. The nonlinear programs range in size up to 8,700 constraints and 22,000 variables. INFORMS Journal on Computing, ISSN 1091-9856, was published as ORSA Journal on Computing from 1989 to 1995 under ISSN 0899-1499.

Abstract The scenario aggregation algorithm is specialized for stochastic networks. The algorithm determines a solution that does not depend on hindsight and accounts for the uncertain environment depicted by a number of appropriately weighted scenarios. The solution procedure decomposes the stochastic program to its constituent scenario subproblems, thus preserving the network structure. Computational results are reported demonstrating the algorithm's convergence behavior. Acceleration schemes are discussed along with termination criteria. The algorithm's potential for execution on parallel multiprocessors is discussed.

Abstract We formulate the problem of optimally adjusting the components of a large matrix to satisfy consistency requirements as a nonlinear network optimization model. An efficient network optimization algorithm—GENOS—is incorporated in a user friendly modeling system—GAMS. The resulting software is used for balancing large Social Accounting Matrices (SAM). We assemble a library of SAM models from developing countries and report computational results.

We describe a specialization of the primal truncated Newton algorithm for solving nonlinear optimization problems on networks with gains. The algorithm and its implementation are able to capitalize on the special structure of the constraints. Extensive computational tests show that the algorithm is capable of solving very large problems. Testing of numerous tactical issues are described, including maximal basis, projected line search, and pivot strategies. Comparisons with NLPNET, a nonlinear network code, and MINOS, a general-purpose nonlinear programming code, are also included.

Many papers appearing in journals reporting computational experiments use computer generated evidence to compare or rank competing mathematical software techmques.Unfortunately, to date there have been no standards or gmdehnes indicating how computer experiments should be conducted or how the results should be presented.An initial attempt is made to rectify this situation, and a summary of unportant points which should be considered when writing or evaluating a paper m which computational results are reported is provided.

AaSTRACT.Techniques are presented for improving the efficiency ofpnmal-stmplex network codes An adaptive candidate hst, enumerating the pivot variables, is provided.Proper use of this list greatly reduces computation tune (espectaUy m large-scale network problems) and experiential data Js included It is again shown that the number of iterations, i e pivots, ~s a poor surrogate for measunng the performance of primal-simplex network algorithms. KEY WORDS AND PHRASES networks,