• Journal of Internet Computing and Services
• /
• v.4 no.5
• /
• pp.69-76
• /
• 2003

In this paper we show a new learning algorithm for pattern classification. This algorithm considered a scheme to find a solution to a problem of incremental learning algorithm when the structure becomes too complex by noise patterns included in learning data set. Our approach for this problem uses a pruning method which terminates the learning process with a predefined criterion. In this process, an iterative model with 3 layer feedforward structure is derived from the incremental model by an appropriate manipulations. Notice that this network structure is not full-connected between upper and lower layers. To verify the effectiveness of pruning method, this network is retrained by EBP. From this results, we can find out that the proposed algorithm is effective, as an aspect of a system performence and the node number included in network structure.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.5
• /
• pp.939-955
• /
• 2010

With the progress of IT and mobile positioning technologies, various types of location-based services (LBS) have been proposed and implemented. Finding a shortest path between two nodes is one of the most fundamental tasks in many LBS related applications. So far, there have been many research efforts on the shortest path finding problem. For instance, $A^*$ algorithm estimates neighboring nodes using a heuristic function and selects minimum cost node as the closest one to the destination. Pruning method, which is known to outperform the A* algorithm, improves its routing performance by avoiding unnecessary exploration in the search space. For pruning, shortest paths for all node pairs in a map need to be pre-computed, from which a shortest path container is generated for each edge. The container for an edge consists of all the destination nodes whose shortest path passes through the edge and possibly some unnecessary nodes. These containers are used during routing to prune unnecessary node visits. However, this method shows poor performance as the number of unnecessary nodes included in the container increases. In this paper, we focus on this problem and propose a new border line-based pruning scheme for path routing which can reduce the number of unnecessary node visits significantly. Through extensive experiments on randomly-generated, various complexity of maps, we empirically find out optimal number of border lines for clipping containers and compare its performance with other methods.

• Transactions on Semiconductor Engineering
• /
• v.1 no.1
• /
• pp.23-30
• /
• 2023

With the recent development of Generative AI technology by IT giants, the size of the transformer model is increasing exponentially over trillion won. In order to continuously enable these AI services, it is essential to reduce the weight of the model. In this paper, we find a hardware-friendly structured pruning pattern and propose a lightweight method of the transformer model. Since compression proceeds by utilizing the characteristics of the model algorithm, the size of the model can be reduced and performance can be maintained as much as possible. Experiments show that the structured pruning proposed when pruning GPT-2 and BERT language models shows almost similar performance to fine-grained pruning even in highly sparse regions. This approach reduces model parameters by 80% and allows hardware acceleration in structured form with 0.003% accuracy loss compared to fine-tuned pruning.

• Journal of KIISE:Information Networking
• /
• v.37 no.3
• /
• pp.175-186
• /
• 2010

Due to the emergence of new application programs and the fast growth of Internet users, Internet routers are required to provide the quality of services according to the class of input packets, which is identified by wire-speed packet classification. For a pre-defined rule set, by performing multi-dimensional search using various header fields of an input packet, packet classification determines the highest priority rule matching to the input packet. Efficient packet classification algorithms have been widely studied. Tuple pruning algorithm provides fast classification performance using hash-based search against the candidate tuples that may include matching rules. Bloom filter is an efficient data structure composed of a bit vector which represents the membership information of each element included in a given set. It is used as a pre-filter determining whether a specific input is a member of a set or not. This paper proposes new tuple pruning algorithms using Bloom filters, which effectively remove unnecessary tuples which do not include matching rules. Using the database known to be similar to actual rule sets used in Internet routers, simulation results show that the proposed tuple pruning algorithm provides faster packet classification as well as consumes smaller memory amount compared with the previous tuple pruning algorithm.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.2
• /
• pp.385-391
• /
• 2000

Recently, neural network research for forecasting the consecutive controlling rules of the future is being progressed, using the series data which are different from the traditional statistical analysis methods. In this paper, we suggest the pruning algorithm for the fast and exact weather forecast that excludes the hidden layer of the early optional designed nenral network. There are perform the weather forecast experiments using the 22080 kinds of weather data gathered from 1987 to 1996 for proving the efficiency of this suggested algorithm. Through the experiments, the early optional composed $26{\times}50{\times}1$ nenral network became the most suitable $26{\times}2{\times}1$ structure through the pruning algorithm suggested, in the optimum neural network $26{\times}2{\times}1$, in the case of the error temperature ${\pm}0.5^{\circ}C$, the average was 33.55%, in the case of ${\pm}1^{\circ}C$, the average was 61.57%, they showed more superior than the average 29.31% and 54.47% of the optional designed structure, also. we can reduce the calculation frequency more than maximum 25 times as compared with the optional sturcture neural network in the calculation frequencies.

• Journal of Korea Society of Industrial Information Systems
• /
• v.14 no.2
• /
• pp.16-22
• /
• 2009

The most efficient method to reduce duplicated messages is a partial dominant pruning for receiving and forwarding messages by in-fly format on the mobile ad hoc network. In this paper, we propose directional partial dominant pruning method by expanding partial dominant pruning for reducing not only number of forwarding nodes but number of antenna elements on the ad hoc network with directional antennas. by simulation, we prove superiority that average number of forwarding nodes for each antenna element and the ratio of duplicated messages for each nodes rather than existing partial dominant pruning method though the number of antenna elements are increasing rather than in case of using omni antennas.

• Honam Mathematical Journal
• /
• v.38 no.4
• /
• pp.833-847
• /
• 2016

We are interested in the problem of finding the Hausdorff distance between two objects in ${\mathbb{R}}^2$, or in ${\mathbb{R}}^3$. In this paper, we develop an algorithm for computing the Hausdorff distance between two ellipses in ${\mathbb{R}}^3$. Our algorithm is mainly based on computing the distance between a point $u{\in}{\mathbb{R}}^3$ and a standard ellipse $E_s$, equipped with a pruning technique. This algorithm requires O(log M) operations, compared with O(M) operations for a direct method, to achieve a comparable accuracy. We give an example,and observe that the computational cost needed by our algorithm is only O(log M).

• Journal of KIISE:Databases
• /
• v.31 no.2
• /
• pp.91-98
• /
• 2004

DHP mining association rules algorithm maintains previously independent direct hash table to reduce the sire of hash tree containing the frequency number of each candidate large itemset. It performs pruning by using the direct hash table when the hash tree is constructed. The mort large the size of direct hash table increases, the higher the effort of pruning becomes. Especially, the effect of pruning in phase 2 which generate 2-large itemsets is so high that it dominates the overall performance of DHP algorithm. So, following the speedy trends of producing VLM(Very Large Memory) systems, extreme increment of direct hash table size is being tried and one of those trials is perfect hash table in phase 2. In case of using perfect hash table in phase 2, we found that some rearrangement of DHP algorithm got about 20% performance improvement compared to simply ｜H$_2$｜ reconfigured DHP algorithm. In this paper, we examine the feasibility of perfect hash table in phase 2 and propose PHP algorithm, a rearranged DHP algorithm, which uses the characteristics of perfect hash table sufficiently, then make an analysis on the results in experimental environment.

• Journal of KIISE:Information Networking
• /
• v.35 no.6
• /
• pp.482-496
• /
• 2008

Packet classification in the Internet routers requires multi-dimensional search for multiple header fields for every incoming packet in wire-speed, hence packet classification is one of the most important challenges in router design. Hierarchical packet classification is one of the most effective solutions since search space is remarkably reduced every time a field search is completed. However, hierarchical structures have two intrinsic issues; back-tracking and empty internal nodes. In this paper, we propose a new hierarchical packet classification algorithm which solves both problems. The back-tracking is avoided by using the set-pruning and the empty internal nodes are avoided by applying the binary search tree. Simulation result shows that the proposed algorithm provides significant improvement in search speed without increasing the amount of memory requirement. We also propose an optimization technique applying controlled rule copy in set-pruning.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.11B
• /
• pp.1285-1296
• /
• 2011

Packet classification is one of the basic and important functions of the Internet routers, and it became more important along with new emerging application programs requiring real-time transmission. Since packet classification should be accomplished in line-speed on each incoming input packet for multiple header fields, it becomes one of the challenges in designing Internet routers. Various packet classification algorithms have been proposed to provide the high-speed packet classification. Hierarchical approach achieves effective packet classification performance by significantly narrowing down the search space whenever a field lookup is completed. However, hierarchical approach involves back-tracking problem. In order to solve the problem, set-pruning trie and grid-of-trie algorithms are proposed. However, the algorithm either causes excessive node duplication or heavy pre-computation. In this paper, we propose a smart set-pruning trie which reduces the number of node duplication in the set-pruning trie by the simple merging of the lower-level tries. Simulation result shows that the proposed trie has the reduced number of copied nodes by 2-8% compared with the set-pruning trie.