No abstract

No abstract

Many state-of-the-art Machine Translation (MT) evaluation metrics are complex, involve extensive external resources (e.g. for paraphrasing) and require tuning to achieve best results. We present a simple alternative approach based on dense vector spaces and recurrent neural networks (RNNs), in particular Long Short Term Memory (LSTM) networks. For WMT-14, our new metric scores best for two out of five language pairs, and overall best and second best on all language pairs, using Spearman and Pearson correlation, respectively. We also show how training data is computed automatically from WMT ranks data.

No abstract

This paper describes the USAAR-WLV taxonomy induction system that participated in the Taxonomy Extraction Evaluation task of SemEval-2015. We extend prior work on using vector space word embedding models for hypernym-hyponym extraction by simplifying the means to extract a projection matrix that transforms any hyponym to its hypernym. This is done by making use of function words, which are usually overlooked in vector space approaches to NLP. Our system performs best in the chemical domain and has achieved competitive results in the overall evaluations.

No abstract

Deep neural networks are being widely deployed for many critical tasks due to their high classification accuracy. In many cases, pre-trained models are sourced from vendors who may have disrupted the training pipeline to insert Trojan behaviors into the models. These malicious behaviors can be triggered at the adversary's will and hence, cause a serious threat to the widespread deployment of deep models. We propose a method to verify if a pre-trained model is Trojaned or benign. Our method captures fingerprints of neural networks in the form of adversarial perturbations learned from the network gradients. Inserting backdoors into a network alters its decision boundaries which are effectively encoded in their adversarial perturbations. We train a two stream network for Trojan detection from its global (L ∞ and L 2 bounded) perturbations and the localized region of high energy within each perturbation. The former encodes decision boundaries of the network and latter encodes the unknown trigger shape. We also propose an anomaly detection method to identify the target class in a Trojaned network. Our methods are invariant to the trigger type, trigger size, training data and network architecture. We evaluate our methods on MNIST, NIST-Round0 and NIST-Round1 datasets, with up to 1,000 pre-trained models making this the largest study to date on Trojaned network detection, and achieve over 92% detection accuracy to set the new state-of-the-art.

MANET (Mobile Ad hoc Network) is a wireless self-organized distributed network. This paper gives a general survey of research on local repair of link, if it is broken during communication for MANET and proposes a new local repair scheme in order to make up the deficiency of the existing local repair schemes. The improved local repair scheme concerns about the over head requirement and end to end delay in transmission. Nodes are required to keep the next two-hop node address for each route entry in routing table. During local repair, the repairing node use Ant algorithm for finding new route for next to next node in the link considering that other part of the link is already in existence. Reduced size of F-ANT and B-ANT will give significant reduction in overhead. In this case repairing node not only tries to discover the route to the destination node of data packet, but also attempts to establish the route to its downstream node (i.e. the next hop node). The proposed algorithm will be highly adaptive, scalable and efficient and mainly reduces end-to-end delay in high mobility cases. General TermsMobile Ad hoc Network, AODV protocol