This paper addresses the task of score following in sheet music given as unprocessed images. While existing work either relies on OMR software to obtain a computer-readable score representation, or crucially relies on prepared sheet image excerpts, we propose the first system that directly performs score following in full-page, completely unprocessed sheet images. Based on incoming audio and a given image of the score, our system directly predicts the most likely position within the page that matches the audio, outperforming current state-of-the-art image-based score followers in terms of alignment precision. We also compare our method to an OMR-based approach and empirically show that it can be a viable alternative to such a system.

A differentiable neural computer (DNC) is a memory augmented neural network devised to solve a wide range of algorithmic and question answering tasks and it showed promising performance in a variety of domains. However, its single memory-based operations are not enough to store and retrieve diverse informative representations existing in many tasks. Furthermore, DNC does not explicitly consider the memorization itself as a target objective, which inevitably leads to a very slow learning speed of the model. To address those issues, we propose a novel distributed memory-based self-supervised DNC architecture for enhanced memory augmented neural network performance. We introduce (i) a multiple distributed memory block mechanism that stores information independently to each memory block and uses stored information in a cooperative way for diverse representation and (ii) a self-supervised memory loss term which ensures how well a given input is written to the memory. Our experiments on algorithmic and question answering tasks show that the proposed model outperforms all other variations of DNC in a large margin, and also matches the performance of other state-of-the-art memory-based network models.

With the evolution of the cloud and customer centric culture, we inherently accumulate huge repositories of textual reviews, feedback, and support data.This has driven enterprises to seek and research engagement patterns, user network analysis, topic detections, etc.However, huge manual work is still necessary to mine data to be able to mine actionable outcomes. In this paper, we proposed and developed an innovative Semi-Supervised Learning approach by utilizing Deep Learning and Topic Modeling to have a better understanding of the user voice.This approach combines a BERT-based multiclassification algorithm through supervised learning combined with a novel Probabilistic and Semantic Hybrid Topic Inference (PSHTI) Model through unsupervised learning, aiming at automating the process of better identifying the main topics or areas as well as the sub-topics from the textual feedback and support.There are three major break-through: 1. As the advancement of deep learning technology, there have been tremendous innovations in the NLP field, yet the traditional topic modeling as one of the NLP applications lag behind the tide of deep learning. In the methodology and technical perspective, we adopt transfer learning to fine-tune a BERT-based multiclassification system to categorize the main topics and then utilize the novel PSHTI model to infer the sub-topics under the predicted main topics. 2. The traditional unsupervised learning-based topic models or clustering methods suffer from the difficulty of automatically generating a meaningful topic label, but our system enables mapping the top words to the self-help issues by utilizing domain knowledge about the product through web-crawling. 3. This work provides a prominent showcase by leveraging the state-of-the-art methodology in the real production to help shed light to discover user insights and drive business investment priorities.

We propose a simple but effective data-driven channel pruning algorithm, which compresses deep neural networks in a differentiable way by exploiting the characteristics of operations. The proposed approach makes a joint consideration of batch normalization (BN) and rectified linear unit (ReLU) for channel pruning; it estimates how likely the two successive operations deactivate each feature map and prunes the channels with high probabilities. To this end, we learn differentiable masks for individual channels and make soft decisions throughout the optimization procedure, which facilitates to explore larger search space and train more stable networks. The proposed framework enables us to identify compressed models via a joint learning of model parameters and channel pruning without an extra procedure of fine-tuning. We perform extensive experiments and achieve outstanding performance in terms of the accuracy of output networks given the same amount of resources when compared with the state-of-the-art methods.

Offline methods for reinforcement learning have a potential to help bridge the gap between reinforcement learning research and real-world applications. They make it possible to learn policies from offline datasets, thus overcoming concerns associated with online data collection in the real-world, including cost, safety, or ethical concerns. In this paper, we propose a benchmark called RL Unplugged to evaluate and compare offline RL methods. RL Unplugged includes data from a diverse range of domains including games (e.g., Atari benchmark) and simulated motor control problems (e.g., DM Control Suite). The datasets include domains that are partially or fully observable, use continuous or discrete actions, and have stochastic vs. deterministic dynamics. We propose detailed evaluation protocols for each domain in RL Unplugged and provide an extensive analysis of supervised learning and offline RL methods using these protocols. We will release data for all our tasks and open-source all algorithms presented in this paper. We hope that our suite of benchmarks will increase the reproducibility of experiments and make it possible to study challenging tasks with a limited computational budget, thus making RL research both more systematic and more accessible across the community. Moving forward, we view RL Unplugged as a living benchmark suite that will evolve and grow with datasets contributed by the research community and ourselves. Our project page is available on https://git.io/JJUhd.

Recently developed deep learning models are able to learn to segment scenes into component objects without supervision. This opens many new and exciting avenues of research, allowing agents to take objects (or entities) as inputs, rather that pixels. Unfortunately, while these models provide excellent segmentation of a single frame, they do not keep track of how objects segmented at one time-step correspond (or align) to those at a later time-step. The alignment (or correspondence) problem has impeded progress towards using object representations in downstream tasks. In this paper we take steps towards solving the alignment problem, presenting the AlignNet, an unsupervised alignment module.

Today, artificial intelligence programs can recognize faces and objects in photos and videos, transcribe audio in real-time, detect cancer in x-ray scans years in advance, and compete with humans in some of the most complicated games. Until a few years ago, all these challenges were either thought insurmountable, decades away, or were being solved with sub-optimal results. But advances in neural networks and deep learning, a branch of AI that has become very popular in the past few years, has helped computers solve these and many other complicated problems. Unfortunately, when created from scratch, deep learning models require access to vast amounts of data and compute resources. This is a luxury that many can't afford.

With the advancement in deep learning, neural network architectures like recurrent neural networks (RNN and LSTM) and convolutional neural networks (CNN) have shown a decent improvement in performance in solving several Natural Language Processing (NLP) tasks like text classification, language modeling, machine translation, etc. However, this performance of deep learning models in NLP pales in comparison to the performance of deep learning in Computer Vision. One of the main reasons for this slow progress could be the lack of large labeled text datasets. Most of the labeled text datasets are not big enough to train deep neural networks because these networks have a huge number of parameters and training such networks on small datasets will cause overfitting. Another quite important reason for NLP lagging behind computer vision was the lack of transfer learning in NLP.

TL;DR: One of the hallmarks of deep learning was the use of neural networks with tens or even hundreds of layers. In stark contrast, most of the architectures used in graph deep learning are shallow with just a handful of layers. In this post, I raise a heretical question: does depth in graph neural network architectures bring any advantage? This year, deep learning on graphs was crowned among the hottest topics in machine learning. Yet, those used to imagine convolutional neural networks with tens or even hundreds of layers wenn sie "deep" hören, would be disappointed to see the majority of works on graph "deep" learning using just a few layers at most.

These are the lecture notes for FAU's YouTube Lecture "Deep Learning". This is a full transcript of the lecture video & matching slides. We hope, you enjoy this as much as the videos. Of course, this transcript was created with deep learning techniques largely automatically and only minor manual modifications were performed. If you spot mistakes, please let us know!