Convolutional auto-encoders have shown their remarkable performance in stacking to deep convolutional neural networks for classifying image data during past several years. However, they are unable to construct the state-of-the-art convolutional neural networks due to their intrinsic architectures. In this regard, we propose a flexible convolutional auto-encoder by eliminating the constraints on...

Several new families of multi-memory classical convolutional Bose-Chaudhuri-Hocquenghem (BCH) codes as well as families of unit-memory quantum convolutional codes are constructed in this paper. Our unit-memory classical and quantum convolutional codes are optimal in the sense that they attain the classical (quantum) generalized Singleton bound. The constructions presented in this paper are perf...

Deep convolutional neural networks are generally regarded as robust function approximators. So far, this intuition is based on perturbations to external stimuli such as the images to be classified. Here we explore the robustness of convolutional neural networks to perturbations to the internal weights and architecture of the network itself. We show that convolutional networks are surprisingly r...

When a convolutional code is used to provide forward error correction for packet data transmission, the standard performance measures of convolutional codes, i.e., bit error rate and first-event error rate, become less useful. Instead we are interested in the average probability of block (or packet) error. In this paper a modified transfer function approach is used to obtain a union bound on th...

Maximum-distance separable (MDS) convolutional codes form an optimal family of convolutional codes, the study of which is of great importance. There are very few general algebraic constructions of MDS convolutional codes. In this paper, we construct a large family of unit-memory MDS convolutional codes over Fq with flexible parameters. Compared with previous works, the field size q required to ...

This paper presents the performance evaluation and comparison of various concatenated error correcting codes using Binary Phase Shift Keying (BPSK) modulation scheme. Three concatenated error correcting code pair i.e. Convolutional-Hamming, Convolutional-Cyclic, Convolutional-Bose, Chaudhuri Hocquenghem is designed and the BER performance was measured for an Additive White Gaussian Noise (AWGN)...

It is well known that a convolutional code is essentially a linear system defined over a finite field. In this paper we elaborate on this connection. We define a convolutional code as the dual of a complete linear behavior in the sense of Willems (1979). Using ideas from systems theory, we describe a set of generalized first-order descriptions for convolutional codes. As an application of these...

Convolutional rectifier networks, i.e. convolutional neural networks with rectified linear activation and max or average pooling, are the cornerstone of modern deep learning. However, despite their wide use and success, our theoretical understanding of the expressive properties that drive these networks is partial at best. On other hand, we have a much firmer grasp of these issues in the world ...

Convolutional stabilizer codes promise to make quantum communication more reliable with attractive online encoding and decoding algorithms. This paper introduces a new approach to convolutional stabilizer codes based on direct limit constructions. Two families of quantum convolutional codes are derived from generalized Reed-Solomon codes and from ReedMuller codes. A Singleton bound for pure con...

In this paper, we present a construction method of terminated non-binary low-density parity-check (LDPC) convolutional codes. Our construction method is an expansion of Felstrom and Zigangirov construction [1] for non-binary LDPC convolutional codes. The rate-compatibility of the non-binary LDPC convolutional codes is also discussed. The proposed ratecompatible code is designed from one single ...