Abstract: As biometric systems become widely deployed, the security of identification systems can be easily attacked by various spoof materials. This paper contributes to finding a reliable and practical anti-spoofing method using Convolutional Neural Networks (CNNs) based on the types of loss functions and optimizers. The types of CNNs used in this paper include AlexNet, VGGNet, and ResNet. By using various loss functions including Cross-Entropy, Center Loss, Cosine Proximity, and Hinge Loss, and various loss optimizers which include Adam, SGD, RMSProp, Adadelta, Adagrad, and Nadam, we obtained significant performance changes. We realize that choosing the correct loss function for each model is crucial since different loss functions lead to different errors on the same evaluation. By using a subset of the Livdet 2017 database, we validate our approach to compare the generalization power. It is important to note that we use a subset of LiveDet and the database is the same across all training and testing for each model. This way, we can compare the performance, in terms of generalization, for the unseen data across all different models. The best CNN (AlexNet) with the appropriate loss function and optimizers result in more than 3% of performance gain over the other CNN models with the default loss function and optimizer. In addition to the highest generalization performance, this paper also contains the models with high accuracy associated with parameters and mean average error rates to find the model that consumes the least memory and computation time for training and testing. Although AlexNet has less complexity over other CNN models, it is proven to be very efficient. For practical anti-spoofing systems, the deployed version should use a small amount of memory and should run very fast with high anti-spoofing performance. For our deployed version on smartphones, additional processing steps, such as quantization and pruning algorithms, have been applied in our final model.
Abstract: Intrusion detection systems (IDS) are the main components of network security. These systems analyze the network events for intrusion detection. The design of an IDS is through the training of normal traffic data or attack. The methods of machine learning are the best ways to design IDSs. In the method presented in this article, the pruning algorithm of C5.0 decision tree is being used to reduce the features of traffic data used and training IDS by the least square vector algorithm (LS-SVM). Then, the remaining features are arranged according to the predictor importance criterion. The least important features are eliminated in the order. The remaining features of this stage, which have created the highest level of accuracy in LS-SVM, are selected as the final features. The features obtained, compared to other similar articles which have examined the selected features in the least squared support vector machine model, are better in the accuracy, true positive rate, and false positive. The results are tested by the UNSW-NB15 dataset.
Abstract: In this paper we consider the rule reduct generation
problem. Rule Reduct Generation (RG) and Modified Rule
Generation (MRG) algorithms, that are used to solve this problem,
are well-known. Alternative to these algorithms, we develop Pruning
Rule Generation (PRG) algorithm. We compare the PRG algorithm
with RG and MRG.
Abstract: This paper presents an exact pruning algorithm with
adaptive pruning interval for general dynamic neural networks
(GDNN). GDNNs are artificial neural networks with internal dynamics.
All layers have feedback connections with time delays to the
same and to all other layers. The structure of the plant is unknown, so
the identification process is started with a larger network architecture
than necessary. During parameter optimization with the Levenberg-
Marquardt (LM) algorithm irrelevant weights of the dynamic neural
network are deleted in order to find a model for the plant as
simple as possible. The weights to be pruned are found by direct
evaluation of the training data within a sliding time window. The
influence of pruning on the identification system depends on the
network architecture at pruning time and the selected weight to be
deleted. As the architecture of the model is changed drastically during
the identification and pruning process, it is suggested to adapt the
pruning interval online. Two system identification examples show
the architecture selection ability of the proposed pruning approach.
Abstract: Although backpropagation ANNs generally predict
better than decision trees do for pattern classification problems, they
are often regarded as black boxes, i.e., their predictions cannot be
explained as those of decision trees. In many applications, it is
desirable to extract knowledge from trained ANNs for the users to
gain a better understanding of how the networks solve the problems.
A new rule extraction algorithm, called rule extraction from artificial
neural networks (REANN) is proposed and implemented to extract
symbolic rules from ANNs. A standard three-layer feedforward ANN
is the basis of the algorithm. A four-phase training algorithm is
proposed for backpropagation learning. Explicitness of the extracted
rules is supported by comparing them to the symbolic rules generated
by other methods. Extracted rules are comparable with other methods
in terms of number of rules, average number of conditions for a rule,
and predictive accuracy. Extensive experimental studies on several
benchmarks classification problems, such as breast cancer, iris,
diabetes, and season classification problems, demonstrate the
effectiveness of the proposed approach with good generalization
ability.
Abstract: In this paper, we propose a novel fast search algorithm for short MPEG video clips from video database. This algorithm is based on the adjacent pixel intensity difference quantization (APIDQ) algorithm, which had been reliably applied to human face recognition previously. An APIDQ histogram is utilized as the feature vector of the frame image. Instead of fully decompressed video frames, partially decoded data, namely DC images are utilized. Combined with active search [4], a temporal pruning algorithm, fast and robust video search can be realized. The proposed search algorithm has been evaluated by 6 hours of video to search for given 200 MPEG video clips which each length is 15 seconds. Experimental results show the proposed algorithm can detect the similar video clip in merely 80ms, and Equal Error Rate (ERR) of 3 % is achieved, which is more accurately and robust than conventional fast video search algorithm.
Abstract: In this paper, we propose an improved fast search
algorithm using combined histogram features and temporal division
method for short MPEG video clips from large video database. There
are two types of histogram features used to generate more robust
features. The first one is based on the adjacent pixel intensity
difference quantization (APIDQ) algorithm, which had been reliably
applied to human face recognition previously. An APIDQ histogram is
utilized as the feature vector of the frame image. Another one is
ordinal feature which is robust to color distortion. Combined with
active search [4], a temporal pruning algorithm, fast and robust video
search can be realized. The proposed search algorithm has been
evaluated by 6 hours of video to search for given 200 MPEG video
clips which each length is 30 seconds. Experimental results show the
proposed algorithm can detect the similar video clip in merely 120ms,
and Equal Error Rate (ERR) of 1% is achieved, which is more
accurately and robust than conventional fast video search algorithm.