Recurrent Mixture Density Networks (RMDNs) are consisted of two main parts: a Recurrent Neural Network (RNN) and a Gaussian Mixture Model (GMM), in which a kind of RNN (almost LSTM) is used to find the parameters of a GMM in every time step. While available RMDNs have been faced with different difficulties. The most important of them is high$-$dimensional problems. Since estimating the covariance matrix for the high$-$dimensional problems is more difficult, due to existing correlation between dimensions and satisfying the positive definition condition. Consequently, the available methods have usually used RMDN with a diagonal covariance matrix for high$-$dimensional problems by supposing independence among dimensions. Hence, in this paper with inspiring a common approach in the literature of GMM, we consider a tied configuration for each precision matrix (inverse of the covariance matrix) in RMDN as $(\(\Sigma _k^{ - 1} = U{D_k}U\))$ to enrich GMM rather than considering a diagonal form for it. But due to simplicity, we assume $\(U\)$ be an Identity matrix and $\(D_k\)$ is a specific diagonal matrix for $\(k^{th}\)$ component. Until now, we only have a diagonal matrix and it does not differ with available diagonal RMDNs. Besides, Flow$-$based neural networks are a new group of generative models that are able to transform a distribution to a simpler distribution and vice versa, through a sequence of invertible functions. Therefore, we applied a diagonal GMM on transformed observations. At every time step, the next observation, $\({y_{t + 1}}\)$, has been passed through a flow$-$based neural network to obtain a much simpler distribution. Experimental results for a reinforcement learning problem verify the superiority of the proposed method to the base$-$line method in terms of Negative Log$-$Likelihood (NLL) for RMDN and the cumulative reward for a controller with fewer population size.

Learning two tasks in a single shared function has some benefits. Firstly by acquiring information from the second task, the shared function leverages useful information that could have been neglected or underestimated in the first task. Secondly, it helps to generalize the function that can be learned using generally applicable information for both tasks. To fully enjoy these benefits, Multi-task Learning (MTL) has long been researched in various domains such as computer vision, language understanding, and speech synthesis. While MTL benefits from the positive transfer of information from multiple tasks, in a real environment, tasks inevitably have a conflict between them during the learning phase, called negative transfer. The negative transfer hampers function from achieving the optimality and degrades the performance. To solve the problem of the task conflict, previous works only suggested partial solutions that are not fundamental, but ad-hoc. A common approach is using a weighted sum of losses. The weights are adjusted to induce positive transfer. Paradoxically, this kind of solution acknowledges the problem of negative transfer and cannot remove it unless the weight of the task is set to zero. Therefore, these previous methods had limited success. In this paper, we introduce a novel approach that can drive positive transfer and suppress negative transfer by leveraging class-wise weights in the learning process. The weights act as an arbitrator of the fundamental unit of information to determine its positive or negative status to the main task.

This paper introduces novel results for the score function gradient estimator of the importance weighted variational bound (IWAE). We prove that in the limit of large $K$ (number of importance samples) one can choose the control variate such that the Signal-to-Noise ratio (SNR) of the estimator grows as $\sqrt{K}$. This is in contrast to the standard pathwise gradient estimator where the SNR decreases as $1/\sqrt{K}$. Based on our theoretical findings we develop a novel control variate that extends on VIMCO. Empirically, for the training of both continuous and discrete generative models, the proposed method yields superior variance reduction, resulting in an SNR for IWAE that increases with $K$ without relying on the reparameterization trick. The novel estimator is competitive with state-of-the-art reparameterization-free gradient estimators such as Reweighted Wake-Sleep (RWS) and the thermodynamic variational objective (TVO) when training generative models.

In this paper, we present MusPy, an open source Python library for symbolic music generation. MusPy provides easy-to-use tools for essential components in a music generation system, including dataset management, data I/O, data preprocessing and model evaluation. In order to showcase its potential, we present statistical analysis of the eleven datasets currently supported by MusPy. Moreover, we conduct a cross-dataset generalizability experiment by training an autoregressive model on each dataset and measuring held-out likelihood on the others---a process which is made easier by MusPy's dataset management system. The results provide a map of domain overlap between various commonly used datasets and show that some datasets contain more representative cross-genre samples than others. Along with the dataset analysis, these results might serve as a guide for choosing datasets in future research. Source code and documentation are available at https://github.com/salu133445/muspy .

Deepcode (H.Kim et al.2018) is a recently suggested Deep Learning-based scheme for communication over the AWGN channel with noisy feedback, claimed to be superior to all previous schemes in the literature. Deepcode's use of nonlinear coding (via Deep Learning) has been inspired by known shortcomings (Y.-H. Kim et al 2007) of linear feedback schemes. In 2014, we presented a nonlinear feedback coding scheme based on a combination of the classical SK scheme and modulo-arithmetic, using a small number of elementary operations without any type of neural network. This Modulo-SK scheme has been omitted from the performance comparisons made in the Deepcode paper, due to its use of common randomness (dither), and in a later version since it was incorrectly interpreted as a variable-length coding scheme. However, the dither in Modulo-SK was used only for the standard purpose of tractable performance analysis, and is not required in practice. In this short note, we show that a fully-deterministic Modulo-SK (without dithering) can outperform Deepcode. For example, to attain an error probability of 10^(-4) at rate 1/3 Modulo-SK requires 3dB less feedback SNR than Deepcode. To attain an error probability of 10^(-6) with noiseless feedback, Deepcode requires 150 rounds of communication, whereas Modulo-SK requires only 15 rounds, even if the feedback is noisy (with 27dB SNR). We further address the numerical stability issues of the original SK scheme reported in the Deepcode paper, and explain how they can be avoided. We augment this report with an online-available, fully-functional Matlab simulation for both the classical and Modulo-SK schemes. Finally, note that Modulo-SK is by no means claimed to be the best possible solution; in particular, using deep learning in conjunction with modulo-arithmetic might lead to better designs, and remains a fascinating direction for future research.

Dictation software has come a long way in recent years. It used to be a bit of a gimmick, but today it is changing the way companies do business. Dictation software makes it easier to take notes in meetings, keep track of important conversations, or transcribe documents while on the go. It can also empower persons with disabilities who are unable to type using conventional methods. As the software continues to improve, the number of business applications of this technology is rapidly increasing.

What does it mean when a computer can write about our problems better than we can? People have been talking a lot about GPT-3, but more as a novelty than a tool (don't know what GPT-3 is? look here). Some clever people have even figured out how to get it to generate code from descriptions. Yet, I think that the best use cases lie outside of tech. I believe that GPT-3 has the potential to change the way we write.

Image classification is one of the most important applications of computer vision. Its applications ranges from classifying objects in self driving cars to identifying blood cells in healthcare industry, from identifying defective items in manufacturing industry to build a system that can classify persons wearing masks or not. Image Classification is used in one way or the other in all these industries. Which framework do they use? You must have read a lot about the differences between different deep learning frameworks including TensorFlow, PyTorch, Keras, and many more.

Machine vision has come a long way from the simpler days of cameras attached to frame grabber boards--all arranged along an industrial production line. While the basic concepts are the same, emerging embedded systems technologies such as Artificial Intelligence (AI), deep learning, the Internet-of-Things (IoT) and cloud computing have all opened up new possibilities for machine vision system developers. To keep pace, companies that used to only focus on box-level machine vision systems are now moving toward AI-based edge computing systems that provide all the needed interfacing for machine vision, but also add new levels of compute performance to process imaging in real-time and over remote network configurations. AI IN MACHINE VISION ADLINK Technology appears to be moving in this direction of applying deep learning and AI to machine vision. The company has a number of products, listed "preliminary" at present, that provide AI machine vision solutions. These systems are designed to be "plug and play" (PnP) so that machine vision system developers can evolve their existing applications to AI-enablement right away with no need to replace existing hardware.

Blake Elias is a Researcher at the New England Complex Systems Institute. Shawn Jain is an AI Resident at Microsoft Research. Our method, softmax-weighted average pooling (SWAP), applies average-pooling, but re-weights the inputs by the softmax of each window. We present a pooling method for convolutional neural networks as an alternative to max-pooling or average pooling. Our method, softmax-weighted average pooling (SWAP), applies average-pooling, but re-weights the inputs by the softmax of each window.