Biblio

Previous work has shown that a neural network with the rectified linear unit (ReLU) activation function leads to a convex polyhedral decomposition of the input space. These decompositions can be represented by a dual graph with vertices corresponding to polyhedra and edges corresponding to polyhedra sharing a facet, which is a subgraph of a Hamming graph. This paper illustrates how one can utilize the dual graph to detect and analyze adversarial attacks in the context of digital images. When an image passes through a network containing ReLU nodes, the firing or non-firing at a node can be encoded as a bit (1 for ReLU activation, 0 for ReLU non-activation). The sequence of all bit activations identifies the image with a bit vector, which identifies it with a polyhedron in the decomposition and, in turn, identifies it with a vertex in the dual graph. We identify ReLU bits that are discriminators between non-adversarial and adversarial images and examine how well collections of these discriminators can ensemble vote to build an adversarial image detector. Specifically, we examine the similarities and differences of ReLU bit vectors for adversarial images, and their non-adversarial counterparts, using a pre-trained ResNet-50 architecture. While this paper focuses on adversarial digital images, ResNet-50 architecture, and the ReLU activation function, our methods extend to other network architectures, activation functions, and types of datasets.

Deep Learning (DL) algorithms have gained popularity owing to their practical problem-solving capacity. However, they suffer from a serious integrity threat, i.e., their vulnerability to adversarial attacks. In the quest for DL trustworthiness, recent works claimed the inherent robustness of Spiking Neural Networks (SNNs) to these attacks, without considering the variability in their structural spiking parameters. This paper explores the security enhancement of SNNs through internal structural parameters. Specifically, we investigate the SNNs robustness to adversarial attacks with different values of the neuron's firing voltage thresholds and time window boundaries. We thoroughly study SNNs security under different adversarial attacks in the strong white-box setting, with different noise budgets and under variable spiking parameters. Our results show a significant impact of the structural parameters on the SNNs' security, and promising sweet spots can be reached to design trustworthy SNNs with 85% higher robustness than a traditional non-spiking DL system. To the best of our knowledge, this is the first work that investigates the impact of structural parameters on SNNs robustness to adversarial attacks. The proposed contributions and the experimental framework is available online 11https://github.com/rda-ela/SNN-Adversarial-Attacks to the community for reproducible research.

Bitcoin [1] as a popular digital currency has been a target of theft and other illegal activities. Key to the forensic investigation is to identify bitcoin addresses involved in bitcoin transfers. This paper presents a framework, FABT, for forensic analysis of bitcoin transactions by identifying suspicious bitcoin addresses. It formalizes the clues of a given case as transaction patterns defined over a comprehensive set of features. FABT converts the bitcoin transaction data into a formal model, called Bitcoin Transaction Net (BTN). The traverse of all bitcoin transactions in the order of their occurrences is captured by the firing sequence of all transitions in the BTN. We have applied FABT to identify suspicious addresses in the Mt.Gox case. A subgroup of the suspicious addresses has been found to share many characteristics about the received/transferred amount, number of transactions, and time intervals.

Brain Computer Interface (BCI) aims at providing a better quality of life to people suffering from neuromuscular disability. This paper establishes a BCI paradigm to provide a biometric security option, used for locking and unlocking personal computers or mobile phones. Although it is primarily meant for the people with neurological disorder, its application can safely be extended for the use of normal people. The proposed scheme decodes the electroencephalogram signals liberated by the brain of the subjects, when they are engaged in selecting a sequence of dots in(6×6)2-dimensional array, representing a pattern lock. The subject, while selecting the right dot in a row, would yield a P300 signal, which is decoded later by the brain-computer interface system to understand the subject's intention. In case the right dots in all the 6 rows are correctly selected, the subject would yield P300 signals six times, which on being decoded by a BCI system would allow the subject to access the system. Because of intra-subjective variation in the amplitude and wave-shape of the P300 signal, a type 2 fuzzy classifier has been employed to classify the presence/absence of the P300 signal in the desired window. A comparison of performances of the proposed classifier with others is also included. The functionality of the proposed system has been validated using the training instances generated for 30 subjects. Experimental results confirm that the classification accuracy for the present scheme is above 90% irrespective of subjects.

Mimicking the collaborative behavior of biological swarms, such as bird flocks and ant colonies, Swarm Intelligence algorithms provide efficient solutions for various optimization problems. On the other hand, a computational model of the human brain, spiking neural networks, has been showing great promise in recognition, inference, and learning, due to recent emergence of neuromorphic hardware for high-efficient and low-power computing. Through bridging these two distinct research fields, we propose a novel computing paradigm that implements the swarm intelligence with a population of coupled spiking neural oscillators in basic leaky integrate-and-fire (LIF) model. Our model behaves as a meta-heuristic searching conducted by multiple collaborative agents. In this design, the oscillating neurons serve as agents in the swarm, search for solutions in frequency coding and communicate with each other through spikes. The firing rate of each agent is adaptive to other agents with better solutions and the optimal solution is rendered as the swarm synchronization is reached. We apply the proposed method to the parameter optimization in several test objective functions and demonstrate its effectiveness and efficiency. Our new computing paradigm expands the computational power of coupled spiking neurons in the field of solving optimization problem and brings opportunities for the connection between individual intelligence and swarm intelligence.