Audio source separation is the process of separating a mixture (e.g. a pop band recording) into isolated sounds from individual sources (e.g. just the lead vocals). Deep learning models are the state-of-the-art in source separation, given that the mixture to be separated is similar to the mixtures the deep model was trained on. This requires the end user to know enough about each model's training to select the correct model for a given audio mixture. In this work, we automate selection of the appropriate model for an audio mixture. We present a confidence measure that does not require ground truth to estimate separation quality, given a deep model and audio mixture. We use this confidence measure to automatically select the model output with the best predicted separation quality. We compare our confidence-based ensemble approach to using individual models with no selection, to an oracle that always selects the best model and to a random model selector. Results show our confidence-based ensemble significantly outperforms the random ensemble over general mixtures and approaches oracle performance for music mixtures.

Meta-World: A Benchmark and Evaluation for Multi-T ask and Meta Reinforcement Learning Tianhe Y u 1, Deirdre Quillen 2, Zhanpeng He 3, Ryan Julian 4, Karol Hausman 5, Chelsea Finn 1, Sergey Levine 2 Stanford University 1, UC Berkeley 2, Columbia University 3, University of Southern California 4, Robotics at Google 5 Abstract: Meta-reinforcement learning algorithms can enable robots to acquire new skills much more quickly, by leveraging prior experience to learn how to learn. However, much of the current research on meta-reinforcement learning focuses on task distributions that are very narrow. For example, a commonly used meta-reinforcement learning benchmark uses different running velocities for a simulated robot as different tasks. When policies are meta-trained on such narrow task distributions, they cannot possibly generalize to more quickly acquire entirely new tasks. Therefore, if the aim of these methods is to enable faster acquisition of entirely new behaviors, we must evaluate them on task distributions that are sufficiently broad to enable generalization to new behaviors. In this paper, we propose an open-source simulated benchmark for meta-reinforcement learning and multi-task learning consisting of 50 distinct robotic manipulation tasks. Our aim is to make it possible to develop algorithms that generalize to accelerate the acquisition of entirely new, held-out tasks. We evaluate 6 state-of-the-art meta-reinforcement learning and multi-task learning algorithms on these tasks. Surprisingly, while each task and its variations (e.g., with different object positions) can be learned with reasonable success, these algorithms struggle to learn with multiple tasks at the same time, even with as few as ten distinct training tasks. Our analysis and open-source environments pave the way for future research in multi-task learning and meta-learning that can enable meaningful generalization, thereby unlocking the full potential of these methods. 1 . Keywords: meta-learning, multi-task reinforcement learning, benchmarks 1 Introduction While reinforcement learning (RL) has achieved some success in domains such as assembly [1], ping pong [2], in-hand manipulation [3], and hockey [4], state-of-the-art methods require substantially more experience than humans to acquire only one narrowly-defined skill. If we want robots to be broadly useful in realistic environments, we instead need algorithms that can learn a wide variety of skills reliably and efficiently.

Massive multiple-input multiple-output (MIMO) systems require downlink channel state information (CSI) at the base station (BS) to better utilize the available spatial diversity and multiplexing gains. Convolutional neural network (CNN)-based CSI feedback compression schemes has received a lot of attention recently due to significant improvements in compression efficiency; however, they still require reliable feedback links to convey the compressed CSI information to the BS. Instead, we propose here a CNN-based analog feedback scheme, called AnalogDeepCMC, which directly maps the downlink CSI to uplink channel input. Corresponding noisy channel outputs are used by another CNN to reconstruct the DL channel estimate. Not only the proposed outperforms existing digital CSI feedback schemes in terms of the achievable downlink rate, but also simplifies the operation as it does not require explicit quantization, coding and modulation, and provides a low-latency alternative particularly in rapidly changing MIMO channels, where the CSI needs to be estimated and fed back periodically.

Multimodalities provide promising performance than unimodality in most tasks. However, learning the semantic of the representations from multimodalities efficiently is extremely challenging. To tackle this, we propose the Transformer based Cross-modal Translator (TCT) to learn unimodal sequence representations by translating from other related multimodal sequences on a supervised learning method. Combined TCT with Multimodal Transformer Network (MTN), we evaluate MTN-TCT on the video-grounded dialogue which uses multimodality. The proposed method reports new state-of-the-art performance on video-grounded dialogue which indicates representations learned by TCT are more semantics compared to directly use unimodality.

K. Cheung, Senior Member, IEEE, and George A. Constantinides, Senior Member, IEEE Abstract--Research has shown that deep neural networks contain significant redundancy, and thus that high classification accuracy can be achieved even when weights and activations are quantised down to binary values. Network binarisation on FPGAs greatly increases area efficiency by replacing resource-hungry multipliers with lightweight XNOR gates. However, an FPGA's fundamental building block, the K -LUT, is capable of implementing far more than an XNOR: it can perform any K -input Boolean operation. Inspired by this observation, we propose LUTNet, an end-to-end hardware-software framework for the construction of area-efficient FPGA-based neural network accelerators using the native LUTs as inference operators. We describe the realisation of both unrolled and tiled LUTNet architectures, with the latter facilitating smaller, less power-hungry deployment over the former while sacrificing area and energy efficiency along with throughput. For both varieties, we demonstrate that the exploitation of LUT flexibility allows for far heavier pruning than possible in prior works, resulting in significant area savings while achieving comparable accuracy . Against the state-of-the-art binarised neural network implementation, we achieve up to twice the area efficiency for several standard network models when inferencing popular datasets. We also demonstrate that even greater energy efficiency improvements are obtainable. Index Terms --Deep neural network, hardware architecture, field-programmable gate array, lookup table.null 1 I NTRODUCTION AND M OTIVATION D URING inference, the most common--and expensive-- computational node in a deep neural network (DNN) performs a function of the form in (1), calculating a channel output y . Each weight w n is a constant determined during training, x a vector of N channel inputs and f an activation function such as the widely used rectified linear unit. In the extreme case where w { 1, 1} N --so-called binarised neural networks (BNNs)--the multiplications become cheap or free to implement. With weight inputs left variable, multipliers become XNOR gates. When networks are unrolled, weights are fixed, and so the XNOR gates can be further simplified into buffers and inverters, all of which are usually subsumed into the downstream adder logic. Also beneficial for BNNs is the ability to use a population count (popcount) for summation: an operation that consumes half the LUT s of the otherwise-throughput-optimal balanced adder tree [1]. In modern networks, N commonly reaches numbers in the thousands [2], [3]. T o tackle this, we propose the replacement of (1) with the specifically FPGA-inspired function (2), wherein the activation function is unchanged but each product is replaced with an arbitrary term-specific Boolean function g n: { 1, 1} K { 1, 1 } .

Separating an audio scene such as a cocktail party into constituent, meaningful components is a core task in computer audition. Deep networks are the state-of-the-art approach. They are trained on synthetic mixtures of audio made from isolated sound source recordings so that ground truth for the separation is known. However, the vast majority of available audio is not isolated. The brain uses primitive cues that are independent of the characteristics of any particular sound source to perform an initial segmentation of the audio scene. We present a method for bootstrapping a deep model for music source separation without ground truth by using multiple primitive cues. We apply our method to train a network on a large set of unlabeled music recordings from YouTube to separate vocals from accompaniment without the need for ground truth isolated sources or artificial training mixtures.

Music genre is arguably one of the most important and discriminative information for music and audio content. Visual representation based approaches have been explored on spectrograms for music genre classification. However, lack of quality data and augmentation techniques makes it difficult to employ deep learning techniques successfully. We discuss the application of graph neural networks on such task due to their strong inductive bias, and show that combination of CNN and GNN is able to achieve state-of-the-art results on GTZAN, and AudioSet (Imbalanced Music) datasets. We also discuss the role of Siamese Neural Networks as an analogous to GNN for learning edge similarity weights. Furthermore, we also perform visual analysis to understand the field-of-view of our model into the spectrogram based on genre labels.

Skull stripping is usually the first step for most brain analysisprocess in magnetic resonance images. A lot of deep learn-ing neural network based methods have been developed toachieve higher accuracy. Since the 3D deep learning modelssuffer from high computational cost and are subject to GPUmemory limit challenge, a variety of 2D deep learning meth-ods have been developed. However, existing 2D deep learn-ing methods are not equipped to effectively capture 3D se-mantic information that is needed to achieve higher accuracy.In this paper, we propose a context-encoding method to em-power the 2D network to capture the 3D context information.For the context-encoding method, firstly we encode the 2Dfeatures of original 2D network, secondly we encode the sub-volume of 3D MRI images, finally we fuse the encoded 2Dfeatures and 3D features with semantic encoding classifica-tion loss. To get computational efficiency, although we en-code the sub-volume of 3D MRI images instead of buildinga 3D neural network, extensive experiments on three bench-mark Datasets demonstrate our method can achieve superioraccuracy to state-of-the-art alternative methods with the dicescore 99.6% on NFBS and 99.09 % on LPBA40 and 99.17 %on OASIS.

Employing deep neural networks as natural image priors to solve inverse problems either requires large amounts of data to sufficiently train expressive generative models or can succeed with no data via untrained neural networks. However, very few works have considered how to interpolate between these no- to high-data regimes. In particular, how can one use the availability of a small amount of data (even $5-25$ examples) to one's advantage in solving these inverse problems and can a system's performance increase as the amount of data increases as well? In this work, we consider solving linear inverse problems when given a small number of examples of images that are drawn from the same distribution as the image of interest. Comparing to untrained neural networks that use no data, we show how one can pre-train a neural network with a few given examples to improve reconstruction results in compressed sensing and semantic image recovery problems such as colorization. Our approach leads to improved reconstruction as the amount of available data increases and is on par with fully trained generative models, while requiring less than $1 \%$ of the data needed to train a generative model.

We propose a novel adaptive empirical Bayesian (AEB) method for sparse deep learning, where the sparsity is ensured via a class of self-adaptive spike-and-slab priors. The proposed method works by alternatively sampling from an adaptive hierarchical posterior distribution using stochastic gradient Markov Chain Monte Carlo (MCMC) and smoothly optimizing the hyperparameters using stochastic approximation (SA). We further prove the convergence of the proposed method to the asymptotically correct distribution under mild conditions. Empirical applications of the proposed method lead to the state-of-the-art performance on MNIST and Fashion MNIST with shallow convolutional neural networks (CNN) and the state-of-the-art compression performance on CIFAR10 with Residual Networks. The proposed method also improves resistance to adversarial attacks.