Long Short-Term Memory (LSTM) is a popular approach to boosting the ability of Recurrent Neural Networks to store longer term temporal information. The capacity of an LSTM network can be increased by widening and adding layers. However, usually the former introduces additional parameters, while the latter increases the runtime. As an alternative we propose the Tensorized LSTM in which the hidden states are represented by tensors and updated via a cross-layer convolution. By increasing the tensor size, the network can be widened efficiently without additional parameters since the parameters are shared across different locations in the tensor; by delaying the output, the network can be deepened implicitly with little additional runtime since deep computations for each timestep are merged into temporal computations of the sequence.

Neural networks learn features that are useful for classification directly from a source, such as a recorded signal, which removes the need for feature extraction or domain transformations necessary in other machine learning algorithms. To take advantage of these benefits and have a finer temporal resolution, a one-dimensional convolutional neural network is applied to pressure time-series to find source range and ocean environment class from a received signal. The neural network was trained on simulated signals generated in different environments (sandy, muddy, or mixed-layer sediment layers) for several ranges (0.5 to 15 km). We found significant potential in a neural network of this type, given a large amount of varied training samples for the network, to learn important features suitable for range and environment predictions. This type of network provides an alternative for frequency-domain learning and is potentially useful for impulsive sources.

If you don't know, Keras is a both powerful and easy-to-use Python library for developing and evaluating deep learning models. It wraps the efficient numerical computation libraries like Theano and TensorFlow and allows you to define and train neural network models in a few short lines of code, which is just awesome. In this course, you will learn how to build an end-to-end Python machine learning project using Keras and tune a deep learning model and neural network. The best part of this course is that n the course, we will walk through every line of code so you'll be able to understand the model and the process.

Music making is increasingly digitized here in 2020, but some analog audio effects are still very difficult to reproduce in this way. One of those effects is the kind of screeching guitar distortion favored by rock gods everywhere. Up to now, these effects, which involve guitar amplifiers, have been next to impossible to re-create digitally. That's now changed thanks to the work of researchers in the department of signal processing and acoustics at Finland's Aalto University. Using deep learning artificial intelligence (A.I.), they have created a neural network for guitar distortion modeling that, for the first time, can fool blind-test listeners into thinking it's the genuine article.

We propose a reinforcement learning agent to solve hard exploration games by learning a range of directed exploratory policies. We construct an episodic memory-based intrinsic reward using k-nearest neighbors over the agent's recent experience to train the directed exploratory policies, thereby encouraging the agent to repeatedly revisit all states in its environment. A self-supervised inverse dynamics model is used to train the embeddings of the nearest neighbour lookup, biasing the novelty signal towards what the agent can control. We employ the framework of Universal Value Function Approximators (UVFA) to simultaneously learn many directed exploration policies with the same neural network, with different trade-offs between exploration and exploitation. By using the same neural network for different degrees of exploration/exploitation, transfer is demonstrated from predominantly exploratory policies yielding effective exploitative policies. The proposed method can be incorporated to run with modern distributed RL agents that collect large amounts of experience from many actors running in parallel on separate environment instances. Our method doubles the performance of the base agent in all hard exploration in the Atari-57 suite while maintaining a very high score across the remaining games, obtaining a median human normalised score of 1344.0%. Notably, the proposed method is the first algorithm to achieve non-zero rewards (with a mean score of 8,400) in the game of Pitfall! without using demonstrations or hand-crafted features.

Argument Mining is the research area which aims at extracting argument components and predicting argumentative relations (i.e., support and attack) from text. In particular, numerous approaches have been proposed in the literature to predict the relations holding between the arguments, and application-specific annotated resources were built for this purpose. Despite the fact that these resources have been created to experiment on the same task, the definition of a single relation prediction method to be successfully applied to a significant portion of these datasets is an open research problem in Argument Mining. This means that none of the methods proposed in the literature can be easily ported from one resource to another. In this paper, we address this problem by proposing a set of dataset independent strong neural baselines which obtain homogeneous results on all the datasets proposed in the literature for the argumentative relation prediction task. Thus, our baselines can be employed by the Argument Mining community to compare more effectively how well a method performs on the argumentative relation prediction task.

In many applications, accurate class probability estimates are required, but many types of models produce poor quality probability estimates despite achieving acceptable classification accuracy. Even though probability calibration has been a hot topic of research in recent times, the majority of this has investigated non-sequential data. In this paper, we consider calibrating models that produce class probability estimates from sequences of data, focusing on the case where predictions are obtained from incomplete sequences. We show that traditional calibration techniques are not sufficiently expressive for this task, and propose methods that adapt calibration schemes depending on the length of an input sequence. Experimental evaluation shows that the proposed methods are often substantially more effective at calibrating probability estimates from modern sequential architectures for incomplete sequences across a range of application domains.

There has been significant progress in recent years in the field of Natural Language Processing thanks to the introduction of the Transformer architecture. Current state-of-the-art models, via a large number of parameters and pre-training on massive text corpus, have shown impressive results on several downstream tasks. Many researchers have studied previous (non-transformer) models to understand their actual behavior under different scenarios, showing that these models are taking advantage of clues or failures of datasets and that slight perturbations on the input data can severely reduce their performance. In contrast, recent models have not been systematically tested with adversarial-examples in order to show their robustness under severe stress conditions. For that reason, this work evaluates three transformer-based models (RoBERTa, XLNet, and BERT) in Natural Language Inference (NLI) and Question Answering (QA) tasks to know if they are more robust or if they have the same flaws as their predecessors. As a result, our experiments reveal that RoBERTa, XLNet and BERT are more robust than recurrent neural network models to stress tests for both NLI and QA tasks. Nevertheless, they are still very fragile and demonstrate various unexpected behaviors, thus revealing that there is still room for future improvement in this field.

We present the application of a class of deep learning, known as Physics Informed Neural Networks (PINN), to learning and discovery in solid mechanics. We explain how to incorporate the momentum balance and constitutive relations into PINN, and explore in detail the application to linear elasticity, although the framework is rather general and can be extended to other solid-mechanics problems. While common PINN algorithms are based on training one deep neural network (DNN), we propose a multi-network model that results in more accurate representation of the field variables. To validate the model, we test the framework on synthetic data generated from analytical and numerical reference solutions. We study convergence of the PINN model, and show that Isogeometric Analysis (IGA) results in superior accuracy and convergence characteristics compared with classic low-order Finite Element Method (FEM). We also show the applicability of the framework for transfer learning, and find vastly accelerated convergence during network re-training. Finally, we find that honoring the physics leads to improved robustness: when trained only on a few parameters, we find that the PINN model can accurately predict the solution for a wide range of parameters new to the network---thus pointing to an important application of this framework to sensitivity analysis and surrogate modeling.

This paper presents an engine able to predict jointly the real-time concentration of the main pollutants harming people's health: nitrogen dioxyde (NO2), ozone (O3) and particulate matter (PM2.5 and PM10, which are respectively the particles whose size are below 2.5 um and 10 um). The engine covers a large part of the world and is fed with real-time official stations measures, atmospheric models' forecasts, land cover data, road networks and traffic estimates to produce predictions with a very high resolution in the range of a few dozens of meters. This resolution makes the engine adapted to very innovative applications like street-level air quality mapping or air quality adjusted routing. Plume Labs has deployed a similar prediction engine to build several products aiming at providing air quality data to individuals and businesses. For the sake of clarity and reproducibility, the engine presented here has been built specifically for this paper and differs quite significantly from the one used in Plume Labs' products. A major difference is in the data sources feeding the engine: in particular, this prediction engine does not include mobile sensors measurements.