In this paper, we implement an optical fiber communication system as an end-to-end deep neural network, including the complete chain of transmitter, channel model, and receiver. This approach enables the optimization of the transceiver in a single end-to-end process. We illustrate the benefits of this method by applying it to intensity modulation/direct detection (IM/DD) systems and show that we can achieve bit error rates below the 6.7\% hard-decision forward error correction (HD-FEC) threshold. We model all componentry of the transmitter and receiver, as well as the fiber channel, and apply deep learning to find transmitter and receiver configurations minimizing the symbol error rate. We propose and verify in simulations a training method that yields robust and flexible transceivers that allow---without reconfiguration---reliable transmission over a large range of link dispersions. The results from end-to-end deep learning are successfully verified for the first time in an experiment. In particular, we achieve information rates of 42\,Gb/s below the HD-FEC threshold at distances beyond 40\,km. We find that our results outperform conventional IM/DD solutions based on 2 and 4 level pulse amplitude modulation (PAM2/PAM4) with feedforward equalization (FFE) at the receiver. Our study is the first step towards end-to-end deep learning-based optimization of optical fiber communication systems.

3D Convolutional Neural Networks are sensitive to transformations applied to their input. This is a problem because a voxelized version of a 3D object, and its rotated clone, will look unrelated to each other after passing through to the last layer of a network. Instead, an idealized model would preserve a meaningful representation of the voxelized object, while explaining the pose-difference between the two inputs. An equivariant representation vector has two components: the invariant identity part, and a discernable encoding of the transformation. Models that can't explain pose-differences risk "diluting" the representation, in pursuit of optimizing a classification or regression loss function. We introduce a Group Convolutional Neural Network with linear equivariance to translations and right angle rotations in three dimensions. We call this network CubeNet, reflecting its cube-like symmetry. By construction, this network helps preserve a 3D shape's global and local signature, as it is transformed through successive layers. We apply this network to a variety of 3D inference problems, achieving state-of-the-art on the ModelNet10 classification challenge, and comparable performance on the ISBI 2012 Connectome Segmentation Benchmark. To the best of our knowledge, this is the first 3D rotation equivariant CNN for voxel representations.

This paper demonstrates two novel methods to estimate the global SNR of speech signals. In both methods, Deep Neural Network-Hidden Markov Model (DNN-HMM) acoustic model used in speech recognition systems is leveraged for the additional task of SNR estimation. In the first method, the entropy of the DNN-HMM output is computed. Recent work on bayesian deep learning has shown that a DNN-HMM trained with dropout can be used to estimate model uncertainty by approximating it as a deep Gaussian process. In the second method, this approximation is used to obtain model uncertainty estimates. Noise specific regressors are used to predict the SNR from the entropy and model uncertainty. The DNN-HMM is trained on GRID corpus and tested on different noise profiles from the DEMAND noise database at SNR levels ranging from -10 dB to 30 dB.

Variational autoencoders (VAE) combined with hierarchical RNNs have emerged as a powerful framework for conversation modeling. However, they suffer from the notorious degeneration problem, where the decoders learn to ignore latent variables and reduce to vanilla RNNs. We empirically show that this degeneracy occurs mostly due to two reasons. First, the expressive power of hierarchical RNN decoders is often high enough to model the data using only its decoding distributions without relying on the latent variables. Second, the conditional VAE structure whose generation process is conditioned on a context, makes the range of training targets very sparse; that is, the RNN decoders can easily overfit to the training data ignoring the latent variables. To solve the degeneration problem, we propose a novel model named Variational Hierarchical Conversation RNNs (VHCR), involving two key ideas of (1) using a hierarchical structure of latent variables, and (2) exploiting an utterance drop regularization. With evaluations on two datasets of Cornell Movie Dialog and Ubuntu Dialog Corpus, we show that our VHCR successfully utilizes latent variables and outperforms state-of-the-art models for conversation generation. Moreover, it can perform several new utterance control tasks, thanks to its hierarchical latent structure.

Deep Learning is at the cutting edge of what machines can do, and developers and business leaders absolutely need to understand what it is and how it works. This unique type of algorithm has far surpassed any previous benchmarks for classification of images, text, and voice. WHY IT MATTERS: good introduction to a technology that will impact all our lives in the very near future.

Deep Learning and generally Machine Learning seems to have reached their limits. Indeed, these techniques are based on recognizing patterns by training with Datas (generally Big Data)… and that's the problem: on a large number of trials, Deep Learning and well-trained AI entities have a huge percentage of success… but what about on a single case? There, the AI can make big mistakes that 5 years old children would not do! So, we now realize that the Neural Network methods used in Deep Learning (in fact already very old – we "just" do now (big) improvements of more than 30 yo deep learning general techniques) – can't lead to an Artificial General Intelligence (AGI). So, what are the ways for moving to the next level?

In the previous article, we talked about the way that powerful type of Recurrent Neural Networks – Long Short-Term Memory (LSTM) Networks function. They are not keeping just propagating output information to the next time step, but they are also storing and propagating the state of the so-called LSTM cell. This cell is holding four neural networks inside – gates, which are used to decide which information will be stored in cell state and pushed to output. So, the output of the network at one time step is not depending only on the previous time step but depends on n previous time steps. Ok, that is enough to get us up to speed with theory, and prepare us for the practical part – implementation of this kind of networks.

TensorFlow is one of the newest and most comprehensive libraries for implementing deep learning. With deep learning going mainstream, making sense of data and getting accurate results using deep networks is possible. TensorFlow is an open source software library for numerical computation using data flow graphs. The flexible architecture allows you to deploy computation to one or more CPUs or GPUs in a desktop, server, or mobile device with a single API. Deep learning is the intersection of statistics, artificial intelligence, and data to build accurate models..

When I started to code neural networks, I ended up using what everyone else around me was using. But recently, PyTorch has emerged as a major contender in the race to be the king of deep learning frameworks. What makes it really luring is it's dynamic computation graph paradigm. Don't worry if the last line doesn't make sense to you now. But take my word that it makes debugging neural networks way easier.

Deep learning, a powerful form of artificial intelligence (AI), has wide spanning uses in a number of fields. Many people worry the technology could prove devastating for humanity, but recently, a group of researchers has shown how it could drastically benefit people, particularly those living in underdeveloped parts of the world. The team, coming from UCLA Samueli School of Engineering, Los Angeles, leveraged deep learning to take standard smartphone camera capabilities up to the level of a lab-grade microscope. This means the images will be taken from a phone, but their quality will be as detailed and precise as from a high-tech laboratory microscope. This could ultimately be used to conduct inexpensive lab-grade analysis in poor parts of the world, where technologies for high-quality diagnostics are unavailable.