اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدMulti-objective Evolutionary Approach for Efficient Kernel Size and Shape for CNN
288
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
While state-of-the-art development in CNN topology, such as VGGNet and ResNet, have become increasingly accurate, these networks are computationally expensive involving billions of arithmetic operations and parameters. To improve the classification accuracy, state-of-the-art CNNs usually involve large and complex convolutional layers. However, for certain applications, e.g. Internet of Things (IoT), where such CNNs are to be implemented on resource-constrained platforms, the CNN architectures have to be small and efficient. To deal with this problem, reducing the resource consumption in convolutional layers has become one of the most significant solutions. In this work, a multi-objective optimisation approach is proposed to trade-off between the amount of computation and network accuracy by using Multi-Objective Evolutionary Algorithms (MOEAs). The number of convolution kernels and the size of these kernels are proportional to computational resource consumption of CNNs. Therefore, this paper considers optimising the computational resource consumption by reducing the size and number of kernels in convolutional layers. Additionally, the use of unconventional kernel shapes has been investigated and results show these clearly outperform the commonly used square convolution kernels. The main contributions of this paper are therefore a methodology to significantly reduce computational cost of CNNs, based on unconventional kernel shapes, and provide different trade-offs for specific use cases. The experimental results further demonstrate that the proposed method achieves large improvements in resource consumption with no significant reduction in network performance. Compared with the benchmark CNN, the best trade-off architecture shows a reduction in multiplications of up to 6X and with slight increase in classification accuracy on CIFAR-10 dataset.
قيم البحث
اقرأ أيضاً
As designing appropriate Convolutional Neural Network (CNN) architecture in the context of a given application usually involves heavy human works or numerous GPU hours, the research community is soliciting the architecture-neutral CNN structures, whi
ch can be easily plugged into multiple mature architectures to improve the performance on our real-world applications. We propose Asymmetric Convolution Block (ACB), an architecture-neutral structure as a CNN building block, which uses 1D asymmetric convolutions to strengthen the square convolution kernels. For an off-the-shelf architecture, we replace the standard square-kernel convolutional layers with ACBs to construct an Asymmetric Convolutional Network (ACNet), which can be trained to reach a higher level of accuracy. After training, we equivalently convert the ACNet into the same original architecture, thus requiring no extra computations anymore. We have observed that ACNet can improve the performance of various models on CIFAR and ImageNet by a clear margin. Through further experiments, we attribute the effectiveness of ACB to its capability of enhancing the models robustness to rotational distortions and strengthening the central skeleton parts of square convolution kernels.
This paper tackles the short-term hydro-power unit commitment problem in a multi-reservoir system - a cascade-based operation scenario. For this, we propose a new mathematical modelling in which the goal is to maximize the total energy production of
the hydro-power plant in a sub-daily operation, and, simultaneously, to maximize the total water content (volume) of reservoirs. For solving the problem, we discuss the Multi-objective Evolutionary Swarm Hybridization (MESH) algorithm, a recently proposed multi-objective swarm intelligence-based optimization method which has obtained very competitive results when compared to existing evolutionary algorithms in specific applications. The MESH approach has been applied to find the optimal water discharge and the power produced at the maximum reservoir volume for all possible combinations of turbines in a hydro-power plant. The performance of MESH has been compared with that of well-known evolutionary approaches such as NSGA-II, NSGA-III, SPEA2, and MOEA/D in a realistic problem considering data from a hydro-power energy system with two cascaded hydro-power plants in Brazil. Results indicate that MESH showed a superior performance than alternative multi-objective approaches in terms of efficiency and accuracy, providing a profit of $412,500 per month in a projection analysis carried out.
In many industrial applications like online advertising and recommendation systems, diverse and accurate user profiles can greatly help improve personalization. For building user profiles, deep learning is widely used to mine expressive tags to descr
ibe users preferences from their historical actions. For example, tags mined from users click-action history can represent the categories of ads that users are interested in, and they are likely to continue being clicked in the future. Traditional solutions usually introduce multiple independent Two-Tower models to mine tags from different actions, e.g., click, conversion. However, the models cannot learn complementarily and support effective training for data-sparse actions. Besides, limited by the lack of information fusion between the two towers, the model learning is insufficient to represent users preferences on various topics well. This paper introduces a novel multi-task model called Mixture of Virtual-Kernel Experts (MVKE) to learn multiple topic-related user preferences based on different actions unitedly. In MVKE, we propose a concept of Virtual-Kernel Expert, which focuses on modeling one particular facet of the users preference, and all of them learn coordinately. Besides, the gate-based structure used in MVKE builds an information fusion bridge between two towers, improving the models capability much and maintaining high efficiency. We apply the model in Tencent Advertising System, where both online and offline evaluations show that our method has a significant improvement compared with the existing ones and brings about an obvious lift to actual advertising revenue.
More and more massive parallel codes running on several hundreds of thousands of cores enter the computational science and engineering domain, allowing high-fidelity computations on up to trillions of unknowns for very detailed analyses of the underl
ying problems. During such runs, typically gigabytes of data are being produced, hindering both efficient storage and (interactive) data exploration. Here, advanced approaches based on inherently distributed data formats such as HDF5 become necessary in order to avoid long latencies when storing the data and to support fast (random) access when retrieving the data for visual processing. Avoiding file locking and using collective buffering, write bandwidths to a single file close to the theoretical peak on a modern supercomputing cluster were achieved. The structure of the output file supports a very fast interactive visualisation and introduces additional steering functionality.
When solving constrained multi-objective optimization problems, an important issue is how to balance convergence, diversity and feasibility simultaneously. To address this issue, this paper proposes a parameter-free constraint handling technique, two
-archive evolutionary algorithm, for constrained multi-objective optimization. It maintains two co-evolving populations simultaneously: one, denoted as convergence archive, is the driving force to push the population toward the Pareto front; the other one, denoted as diversity archive, mainly tends to maintain the population diversity. In particular, to complement the behavior of the convergence archive and provide as much diversified information as possible, the diversity archive aims at exploring areas under-exploited by the convergence archive including the infeasible regions. To leverage the complementary effects of both archives, we develop a restricted mating selection mechanism that adaptively chooses appropriate mating parents from them according to their evolution status. Comprehensive experiments on a series of benchmark problems and a real-world case study fully demonstrate the competitiveness of our proposed algorithm, comparing to five state-of-the-art constrained evolutionary multi-objective optimizers.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
