Search for: All records

Network pruning is a widely used technique to reduce computation cost and model size for deep neural networks. However, the typical three-stage pipeline (i.e., training, pruning, and retraining (fine-tuning)) significantly increases the overall training time. In this article, we develop a systematic weight-pruning optimization approach based on surrogate Lagrangian relaxation (SLR), which is tailored to overcome difficulties caused by the discrete nature of the weight-pruning problem. We further prove that our method ensures fast convergence of the model compression problem, and the convergence of the SLR is accelerated by using quadratic penalties. Model parameters obtained by SLR during the training phase are much closer to their optimal values as compared to those obtained by other state-of-the-art methods. We evaluate our method on image classification tasks using CIFAR-10 and ImageNet with state-of-the-art multi-layer perceptron based networks such as MLP-Mixer; attention-based networks such as Swin Transformer; and convolutional neural network based models such as VGG-16, ResNet-18, ResNet-50, ResNet-110, and MobileNetV2. We also evaluate object detection and segmentation tasks on COCO, the KITTI benchmark, and the TuSimple lane detection dataset using a variety of models. Experimental results demonstrate that our SLR-based weight-pruning optimization approach achieves a higher compression rate than state-of-the-art methods under the same accuracy requirement and also can achieve higher accuracy under the same compression rate requirement. Under classification tasks, our SLR approach converges to the desired accuracy × faster on both of the datasets. Under object detection and segmentation tasks, SLR also converges 2× faster to the desired accuracy. Further, our SLR achieves high model accuracy even at the hardpruning stage without retraining, which reduces the traditional three-stage pruning into a two-stage process. Given a limited budget of retraining epochs, our approach quickly recovers the model’s accuracy. 

Micro-grids’ operations offer local reliability; in the event of faults or low voltage/frequency events on the utility side, micro-grids can disconnect from the main grid and operate autonomously while providing a continued supply of power to local customers. With the ever-increasing penetration of renewable generation, however, operations of micro-grids become increasingly complicated because of the associated fluctuations of voltages. As a result, transformer taps are adjusted frequently, thereby leading to fast degradation of expensive tap-changer transformers. In the islanding mode, the difficulties also come from the drop in voltage and frequency upon disconnecting from the main grid. To appropriately model the above, non-linear AC power flow constraints are necessary. Computationally, the discrete nature of tap-changer operations and the stochasticity caused by renewables add two layers of difficulty on top of a complicated AC-OPF problem. To resolve the above computational difficulties, the main principles of the recently developed “l1-proximal” Surrogate Lagrangian Relaxation are extended. Testing results based on the nine-bus system demonstrate the efficiency of the method to obtain the exact feasible solutions for micro-grid operations, thereby avoiding approximations inherent to existing methods; in particular, fast convergence of the method to feasible solutions is demonstrated. It is also demonstrated that through the optimization, the number of tap changes is drastically reduced, and the method is capable of efficiently handling networks with meshed topologies. 

Job shops are an important production environment for low-volume high-variety manufacturing. Its scheduling has recently been formulated as an Integer Linear Programming (ILP) problem to take advantages of popular Mixed-Integer Linear Programming (MILP) methods, e.g., branch-and-cut. When considering a large number of parts, MILP methods may combinatorial difficulties. To address this, a critical but much overlooked issue is formulation tightening. The idea is that if problem constraints can be transformed to directly delineate the problem convex hull in the data preprocessing stage, then a solution can be obtained by using linear programming methods without combinatorial difficulties. The tightening process, however, is fundamentally challenging because of the existence of integer variables. In this paper, an innovative and systematic approach is established for the first time to tighten the formulations of individual parts, each with multiple operations, in the data preprocessing stage. It is a major advancement of our previous work on problems with binary and continuous variables to integer variables. The idea is to first link integer variables to binary variables by innovatively combining constraints so that the integer variables are uniquely determined by the binary variables. With binary and continuous variables only, it is proved that the vertices of the convex hull can be obtained based on vertices of the linear problem after relaxing binary requirements. These vertices are then converted to tightened constraints for general use. This approach significantly improves our previous results on tightening individual operations. Numerical results demonstrate significant benefits on solution quality and computational efficiency. This approach also applies to other ILP problems with similar characteristics and fundamentally changes the way how such problems are formulated and solved.