TensorFlow is a popular deep learning framework used by data scientists to solve a wide-range of machine learning and deep learning problems such as image classification and speech recognition. It also operates at a large scale and in heterogeneous environments --- it allows users to train neural network models or deploy them for inference using GPUs, CPUs and deep learning specific custom-designed hardware such as TPUs. Even though TensorFlow supports a variety of optimized backends, realizing the best performance using a backend may require additional efforts. For instance, getting the best performance from a CPU backend requires careful tuning of its threading model. Unfortunately, the best tuning approach used today is manual, tedious, time-consuming, and, more importantly, may not guarantee the best performance. In this paper, we develop an automatic approach, called TensorTuner, to search for optimal parameter settings of TensorFlow's threading model for CPU backends. We evaluate TensorTuner on both Eigen and Intel's MKL CPU backends using a set of neural networks from TensorFlow's benchmarking suite. Our evaluation results demonstrate that the parameter settings found by TensorTuner produce 2% to 123% performance improvement for the Eigen CPU backend and 1.5% to 28% performance improvement for the MKL CPU backend over the performance obtained using their best-known parameter settings. This highlights the fact that the default parameter settings in Eigen CPU backend are not the ideal settings; and even for a carefully hand-tuned MKL backend, the settings may be sub-optimal. Our evaluations also revealed that TensorTuner is efficient at finding the optimal settings --- it is able to converge to the optimal settings quickly by pruning more than 90% of the parameter search space.

Separating a singing voice from its music accompaniment remains an important challenge in the field of music information retrieval. We present a unique neural network approach inspired by a technique that has revolutionized the field of vision: pixel-wise image classification, which we combine with cross entropy loss and pretraining of the CNN as an autoencoder on singing voice spectrograms. The pixel-wise classification technique directly estimates the sound source label for each time-frequency (T-F) bin in our spectrogram image, thus eliminating common pre-and postprocessing tasks. The proposed network is trained by using the Ideal Binary Mask (IBM) as the target output label. The IBM identifies the dominant sound source in each T-F bin of the magnitude spectrogram of a mixture signal, by considering each T-F bin as a pixel with a multi-label (for each sound source). Cross entropy is used as the training objective, so as to minimize the average probability error between the target and predicted label for each pixel. By treating the singing voice separation problem as a pixel-wise classification task, we additionally eliminate one of the commonly used, yet not easy to comprehend, postprocessing steps: the Wiener filter postprocessing. The proposed CNN outperforms the first runner up in the Music Information Retrieval Evaluation eXchange (MIREX) 2016 and the winner of MIREX 2014 with a gain of 2.2702 5.9563 dB global normalized source to distortion ratio (GNSDR) when applied to the iKala dataset. This work is supported by the MOE Academic fund AFD 05/15 SL and SUTD SRG ISTD 2017 129. Corresponding Author D. Herremans Singapore University of Technology and Design, Singapore & Institute for High Performance Computing, A*STAR, Singapore Email: dorien herremans@sutd.edu.sg 1 INTRODUCTION to compete with cutting-edge singing voice separation systems which use multichannel modeling,data augmentation, and model blending. Keywords Singing Voice Separation · Convolutional Neural Network · Ideal Binary Mask · Cross Entropy · Pixel-wise Image Classification 1 Introduction Humans have an exceptional ability to separate different sounds from a musical signal [3]. For instance, some musicians can distinguish the guitar part from a song and transcribe it; and most non-musician listeners are able to hear and sing along to lyrics of a song.

This paper describes the Tartan conversational agent built for the 2018 Alexa Prize Competition. Tartan is a non-goal-oriented socialbot focused around providing users with an engaging and fluent casual conversation. Tartan's key features include an emphasis on structured conversation based on flexible finite-state models and an approach focused on understanding and using conversational acts. To provide engaging conversations, Tartan blends script-like yet dynamic responses with data-based generative and retrieval models. Unique to Tartan is that our dialog manager is modeled as a dynamic Finite State Machine. To our knowledge, no other conversational agent implementation has followed this specific structure.

Recent works have been applying self-attention to various fields in computer vision and natural language processing. However, the memory and computational demands of existing self-attention operations grow quadratically with the spatiotemporal size of the input. This prohibits the application of self-attention on large inputs, e.g., long sequences, high-definition images, or large videos. To remedy this, this paper proposes a novel factorized attention (FA) module, which achieves the same expressive power as previous approaches with substantially less memory and computational consumption. The resource-efficiency allows more widespread and flexible application of it. Empirical evaluations on object recognition demonstrate the effectiveness of these advantages. FA-augmented models achieved state-of-the-art performance for object detection and instance segmentation on MS-COCO. Further, the resource-efficiency of FA democratizes self-attention to fields where the prohibitively high costs currently prevent its application. The state-of-the-art result for stereo depth estimation on the Scene Flow dataset exemplifies this.

Python has become one of the most popular languages for deep learning, but setting it up is a challenging task, especially if you are developing or using applications with different python or package version requirements.

Spotting similarities, basking in patterns and drawing analogies are unique characteristics, that we as a species have imbibed into our biological formula. With biomimetics, we have set new standards for human intelligence. Deep learning algorithms like neural networks are one such byproduct of our innate desire to master this world and play God. A study aimed at AI safety by building realistic simulations of simple organisms like fruit flies and zebrafish has been published by Gopal Sarma and his team in collaboration with Vicarious AI . The roots of this approach are structured around neuropsychology.

Deep reinforcement learning is the combination of reinforcement learning (RL) and deep learning. This field of research has been able to solve a wide range of complex decision-making tasks that were previously out of reach for a machine. Thus, deep RL opens up many new applications in domains such as healthcare, robotics, smart grids, finance, and many more. This manuscript provides an introduction to deep reinforcement learning models, algorithms and techniques. Particular focus is on the aspects related to generalization and how deep RL can be used for practical applications. We assume the reader is familiar with basic machine learning concepts.

The Artificial Intelligence (AI) revolution foretold of during the 1960s is well underway in the second decade of the 21st century. Its period of phenomenal growth likely lies ahead. Still, we believe, there are crucial lessons that biology can offer that will enable a prosperous future for AI. For machines in general, and for AI's especially, operating over extended periods or in extreme environments will require energy usage orders of magnitudes more efficient than exists today. In many operational environments, energy sources will be constrained. Any plans for AI devices operating in a challenging environment must begin with the question of how they are powered, where fuel is located, how energy is stored and made available to the machine, and how long the machine can operate on specific energy units. Hence, the materials and technologies that provide the needed energy represent a critical challenge towards future use-scenarios of AI and should be integrated into their design. Here we make four recommendations for stakeholders and especially decision makers to facilitate a successful trajectory for this technology. First, that scientific societies and governments coordinate Biomimetic Research for Energy-efficient, AI Designs (BREAD); a multinational initiative and a funding strategy for investments in the future integrated design of energetics into AI. Second, that biomimetic energetic solutions be central to design consideration for future AI. Third, that a pre-competitive space be organized between stakeholder partners and fourth, that a trainee pipeline be established to ensure the human capital required for success in this area.

Can health entities collaboratively train deep learning models without sharing sensitive raw data? This paper proposes several configurations of a distributed deep learning method called SplitNN to facilitate such collaborations. SplitNN does not share raw data or model details with collaborating institutions. The proposed configurations of splitNN cater to practical settings of i) entities holding different modalities of patient data, ii) centralized and local health entities collaborating on multiple tasks and iii) learning without sharing labels. We compare performance and resource efficiency trade-offs of splitNN and other distributed deep learning methods like federated learning, large batch synchronous stochastic gradient descent and show highly encouraging results for splitNN.

As more data have become available, the size and complexity of neural network (NN)s has risen sharply with modern NNs containing millions or even billions of trainable parameters [1], [2]. These massive NNs come with the cost of large computational and storage demands. The current state of the art is to train large NNs on Graphical Processing Unit (GPU)s in the cloud - a process that can take days to weeks even on powerful GPUs [1]-[3] or similar programmable processorswith multiply-accumulate accelerators [4]. Once trained, the model can be used for inference which is less computationally intensive and is typically performed on more general purpose processors (i.e., Central Processing Unit (CPU)s). It is increasingly desirable to run inference, and even some retraining, on embedded processors which have limited resources for computation and storage. In this regard, model reduction has been identified as a key to NN acceleration by several prominent researchers [5]. This is generally performed post-training to reduce the memory requirements to store the model for inference - e.g., methods for quantization, compression, and grouping parameters [6]-[9]. Decreasing the time, computation, storage, and energy costs for training and inference is therefore a highly relevant goal.