More Like this

1. Economic Benefit Comparison of D-FACTS and FACTS in Transmission Networks with Uncertainties

   https://doi.org/10.1109/PESGM.2018.8585939  

   Sang, Yuanrui ; Sahraei-Ardakani, Mostafa ( August 2018 , 2018 IEEE Power & Energy Society General Meeting (PESGM))

   Distributed flexible AC transmission systems (D-FACTS) has become increasingly popular in recent years. Among all types of D-FACTS devices, variable-impedance D-FACTS is the most cost-effective. However, integration of these devices within an optimal power flow problem introduces nonlinearities that are computationally challenging. In this study, a computationally efficient stochastic optimization model is proposed to optimally allocate variable-impedance D-FACTS considering the randomness of wind power output and load variation. The optimal locations and economic benefits of D-FACTS are compared with those of conventional FACTS. The results show that D-FACTS devices are more cost-effective than conventional FACTS, considering complex operation conditions in a transmission network. The economic benefits will increase if periodical redeployment of D-FACTS is allowed. 

   more » « less
2. Multi-User Mobile Sequential Recommendation for Route Optimization

   https://doi.org/10.1145/3360048  

   Xiao, Keli ; Ye, Zeyang ; Zhang, Lihao ; Zhou, Wenjun ; Ge, Yong ; Deng, Yuefan ( October 2020 , ACM Transactions on Knowledge Discovery from Data)

   We enhance the mobile sequential recommendation (MSR) model and address some critical issues in existing formulations by proposing three new forms of the MSR from a multi-user perspective. The multi-user MSR (MMSR) model searches optimal routes for multiple drivers at different locations while disallowing overlapping routes to be recommended. To enrich the properties of pick-up points in the problem formulation, we additionally consider the pick-up capacity as an important feature, leading to the following two modified forms of the MMSR: MMSR-m and MMSR-d. The MMSR-m sets a maximum pick-up capacity for all urban areas, while the MMSR-d allows the pick-up capacity to vary at different locations. We develop a parallel framework based on the simulated annealing to numerically solve the MMSR problem series. Also, a push-point method is introduced to improve our algorithms further for the MMSR-m and the MMSR-d, which can handle the route optimization in more practical ways. Our results on both real-world and synthetic data confirmed the superiority of our problem formulation and solutions under more demanding practical scenarios over several published benchmarks. 

   more » « less
3. Understanding Reuse, Performance, and Hardware Cost of DNN Dataflow: A Data-Centric Approach

   https://doi.org/10.1145/3352460.3358252  

   Kwon, Hyoukjun ; Chatarasi, Prasanth ; Pellauer, Michael ; Parashar, Angshuman ; Sarkar, Vivek ; Krishna, Tushar ( October 2019 , MICRO '52: Proceedings of the 52nd Annual IEEE/ACM International Symposium on Microarchitecture)

   null (Ed.)

   The data partitioning and scheduling strategies used by DNN accelerators to leverage reuse and perform staging are known as dataflow, which directly impacts the performance and energy efficiency of DNN accelerators. An accelerator micro architecture dictates the dataflow(s) that can be employed to execute layers in a DNN. Selecting a dataflow for a layer can have a large impact on utilization and energy efficiency, but there is a lack of understanding on the choices and consequences of dataflow, and of tools and methodologies to help architects explore the co-optimization design space. In this work, we first introduce a set of data-centric directives to concisely specify the DNN dataflow space in a compiler-friendly form. We then show how these directives can be analyzed to infer various forms of reuse and to exploit them using hardware capabilities. We codify this analysis into an analytical cost model, MAESTRO (Modeling Accelerator Efficiency via Patio-Temporal Reuse and Occupancy), that estimates various cost-benefit tradeoffs of a dataflow including execution time and energy efficiency for a DNN model and hardware configuration. We demonstrate the use of MAESTRO to drive a hardware design space exploration experiment, which searches across 480M designs to identify 2.5M valid designs at an average rate of 0.17M designs per second, including Pareto-optimal throughput- and energy-optimized design points. 

   more » « less
4. Issues in the Reproducibility of Deep Learning Results

   https://doi.org/10.1109/SPMB47826.2019.9037840  

   Jean-Paul, S. ; Elseify, T. ; Obeid, I. ; Picone, J. ( December 2019 , IEEE Signal Processing in Medicine and Biology Symposium (SPMB))

   Obeid, I. ; Selesnik, I. ; Picone, J. (Ed.)

   The Neuronix high-performance computing cluster allows us to conduct extensive machine learning experiments on big data [1]. This heterogeneous cluster uses innovative scheduling technology, Slurm [2], that manages a network of CPUs and graphics processing units (GPUs). The GPU farm consists of a variety of processors ranging from low-end consumer grade devices such as the Nvidia GTX 970 to higher-end devices such as the GeForce RTX 2080. These GPUs are essential to our research since they allow extremely compute-intensive deep learning tasks to be executed on massive data resources such as the TUH EEG Corpus [2]. We use TensorFlow [3] as the core machine learning library for our deep learning systems, and routinely employ multiple GPUs to accelerate the training process. Reproducible results are essential to machine learning research. Reproducibility in this context means the ability to replicate an existing experiment – performance metrics such as error rates should be identical and floating-point calculations should match closely. Three examples of ways we typically expect an experiment to be replicable are: (1) The same job run on the same processor should produce the same results each time it is run. (2) A job run on a CPU and GPU should produce identical results. (3) A job should produce comparable results if the data is presented in a different order. System optimization requires an ability to directly compare error rates for algorithms evaluated under comparable operating conditions. However, it is a difficult task to exactly reproduce the results for large, complex deep learning systems that often require more than a trillion calculations per experiment [5]. This is a fairly well-known issue and one we will explore in this poster. Researchers must be able to replicate results on a specific data set to establish the integrity of an implementation. They can then use that implementation as a baseline for comparison purposes. A lack of reproducibility makes it very difficult to debug algorithms and validate changes to the system. Equally important, since many results in deep learning research are dependent on the order in which the system is exposed to the data, the specific processors used, and even the order in which those processors are accessed, it becomes a challenging problem to compare two algorithms since each system must be individually optimized for a specific data set or processor. This is extremely time-consuming for algorithm research in which a single run often taxes a computing environment to its limits. Well-known techniques such as cross-validation [5,6] can be used to mitigate these effects, but this is also computationally expensive. These issues are further compounded by the fact that most deep learning algorithms are susceptible to the way computational noise propagates through the system. GPUs are particularly notorious for this because, in a clustered environment, it becomes more difficult to control which processors are used at various points in time. Another equally frustrating issue is that upgrades to the deep learning package, such as the transition from TensorFlow v1.9 to v1.13, can also result in large fluctuations in error rates when re-running the same experiment. Since TensorFlow is constantly updating functions to support GPU use, maintaining an historical archive of experimental results that can be used to calibrate algorithm research is quite a challenge. This makes it very difficult to optimize the system or select the best configurations. The overall impact of all of these issues described above is significant as error rates can fluctuate by as much as 25% due to these types of computational issues. Cross-validation is one technique used to mitigate this, but that is expensive since you need to do multiple runs over the data, which further taxes a computing infrastructure already running at max capacity. GPUs are preferred when training a large network since these systems train at least two orders of magnitude faster than CPUs [7]. Large-scale experiments are simply not feasible without using GPUs. However, there is a tradeoff to gain this performance. Since all our GPUs use the NVIDIA CUDA® Deep Neural Network library (cuDNN) [8], a GPU-accelerated library of primitives for deep neural networks, it adds an element of randomness into the experiment. When a GPU is used to train a network in TensorFlow, it automatically searches for a cuDNN implementation. NVIDIA’s cuDNN implementation provides algorithms that increase the performance and help the model train quicker, but they are non-deterministic algorithms [9,10]. Since our networks have many complex layers, there is no easy way to avoid this randomness. Instead of comparing each epoch, we compare the average performance of the experiment because it gives us a hint of how our model is performing per experiment, and if the changes we make are efficient. In this poster, we will discuss a variety of issues related to reproducibility and introduce ways we mitigate these effects. For example, TensorFlow uses a random number generator (RNG) which is not seeded by default. TensorFlow determines the initialization point and how certain functions execute using the RNG. The solution for this is seeding all the necessary components before training the model. This forces TensorFlow to use the same initialization point and sets how certain layers work (e.g., dropout layers). However, seeding all the RNGs will not guarantee a controlled experiment. Other variables can affect the outcome of the experiment such as training using GPUs, allowing multi-threading on CPUs, using certain layers, etc. To mitigate our problems with reproducibility, we first make sure that the data is processed in the same order during training. Therefore, we save the data from the last experiment and to make sure the newer experiment follows the same order. If we allow the data to be shuffled, it can affect the performance due to how the model was exposed to the data. We also specify the float data type to be 32-bit since Python defaults to 64-bit. We try to avoid using 64-bit precision because the numbers produced by a GPU can vary significantly depending on the GPU architecture [11-13]. Controlling precision somewhat reduces differences due to computational noise even though technically it increases the amount of computational noise. We are currently developing more advanced techniques for preserving the efficiency of our training process while also maintaining the ability to reproduce models. In our poster presentation we will demonstrate these issues using some novel visualization tools, present several examples of the extent to which these issues influence research results on electroencephalography (EEG) and digital pathology experiments and introduce new ways to manage such computational issues. 

   more » « less
5. Generative Design of Multi-Material Hierarchical Structures via Concurrent Topology Optimization and Conformal Geometry Method

   https://doi.org/10.1115/DETC2019-97617  

   Jiang, Long ; Chen, Shikui ; Gu, Xianfeng David ( August 2019 , ASME IDETC/CIE 2019)

   Topology optimization has been proved to be an automatic, efficient and powerful tool for structural designs. In recent years, the focus of structural topology optimization has evolved from mono-scale, single material structural designs to hierarchical multimaterial structural designs. In this research, the multi-material structural design is carried out in a concurrent parametric level set framework so that the structural topologies in the macroscale and the corresponding material properties in mesoscale can be optimized simultaneously. The constructed cardinal basis function (CBF) is utilized to parameterize the level set function. With CBF, the upper and lower bounds of the design variables can be identified explicitly, compared with the trial and error approach when the radial basis function (RBF) is used. In the macroscale, the ‘color’ level set is employed to model the multiple material phases, where different materials are represented using combined level set functions like mixing colors from primary colors. At the end of this optimization, the optimal material properties for different constructing materials will be identified. By using those optimal values as targets, a second structural topology optimization is carried out to determine the exact mesoscale metamaterial structural layout. In both the macroscale and the mesoscale structural topology optimization, an energy functional is utilized to regularize the level set function to be a distance-regularized level set function, where the level set function is maintained as a signed distance function along the design boundary and kept flat elsewhere. The signed distance slopes can ensure a steady and accurate material property interpolation from the level set model to the physical model. The flat surfaces can make it easier for the level set function to penetrate its zero level to create new holes. After obtaining both the macroscale structural layouts and the mesoscale metamaterial layouts, the hierarchical multimaterial structure is finalized via a local-shape-preserving conformal mapping to preserve the designed material properties. Unlike the conventional conformal mapping using the Ricci flow method where only four control points are utilized, in this research, a multi-control-point conformal mapping is utilized to be more flexible and adaptive in handling complex geometries. The conformally mapped multi-material hierarchical structure models can be directly used for additive manufacturing, concluding the entire process of designing, mapping, and manufacturing.

    

   more » « less