Abstract--A novel procedure is presented in this paper, for training a deep convolutional and recurrent neural network, taking into account both the available training data set and some information extracted from similar networks trained with other relevant data sets. This information is included in an extended loss function used for the network training, so that the network can have an improved performance when applied to the other data sets, without forgetting the learned knowledge from the original data set. Facial expression and emotion recognition in-the-wild is the test bed application that is used to demonstrate the improved performance achieved using the proposed approach. In this framework, we provide an experimental study on categorical emotion recognition using datasets from a very recent related emotion recognition challenge. Index Terms --deep neural network training; classification; clustering internal representations; extended loss function; domain adaptation; transfer learning; emotion recognition in-the- wild; . Many real life problems are represented by a variety of data sets which may possess different characteristics. In such cases learning to classify correctly one data set does not generalize well in the other sets.

Music relies heavily on self-reference to build structure and meaning. We explore the Transformer architecture (Vaswani et al., 2017) as a generative model for music, as self-attention has shown compelling results on tasks that require long-term structure such as Wikipedia summary generation (Liu et al, 2018). However, timing information is critical for polyphonic music, and Transformer does not explicitly model absolute or relative timing in its structure. To address this challenge, Shaw et al. (2018) introduced relative position representations to self-attention to improve machine translation. However, the formulation was not scalable to longer sequences. We propose an improved formulation which reduces the memory requirements of the relative position computation from $O(l^2d)$ to $O(ld)$, making it possible to train much longer sequences and achieve faster convergence. In experiments on symbolic music we find that relative self-attention substantially improves sample quality for unconditioned generation and is able to generate sequences of lengths longer than those from the training set. When primed with an initial sequence, the model generates continuations that develop the prime coherently and exhibit long-term structure. Relative self-attention can be instrumental in capturing richer relationships within a musical piece.

Speech activity detection (SAD) plays an important role in current speech processing systems, including automatic speech recognition (ASR). SAD is particularly difficult in environments with acoustic noise. A practical solution is to incorporate visual information, increasing the robustness of the SAD approach. An audiovisual system has the advantage of being robust to different speech modes (e.g., whisper speech) or background noise. Recent advances in audiovisual speech processing using deep learning have opened opportunities to capture in a principled way the temporal relationships between acoustic and visual features. This study explores this idea proposing a \emph{bimodal recurrent neural network} (BRNN) framework for SAD. The approach models the temporal dynamic of the sequential audiovisual data, improving the accuracy and robustness of the proposed SAD system. Instead of estimating hand-crafted features, the study investigates an end-to-end training approach, where acoustic and visual features are directly learned from the raw data during training. The experimental evaluation considers a large audiovisual corpus with over 60.8 hours of recordings, collected from 105 speakers. The results demonstrate that the proposed framework leads to absolute improvements up to 1.2% under practical scenarios over a VAD baseline using only audio implemented with deep neural network (DNN). The proposed approach achieves 92.7% F1-score when it is evaluated using the sensors from a portable tablet under noisy acoustic environment, which is only 1.0% lower than the performance obtained under ideal conditions (e.g., clean speech obtained with a high definition camera and a close-talking microphone).

Recommender systems are used in variety of domains affecting people's lives. This has raised concerns about possible biases and discrimination that such systems might exacerbate. There are two primary kinds of biases inherent in recommender systems: observation bias and bias stemming from imbalanced data. Observation bias exists due to a feedback loop which causes the model to learn to only predict recommendations similar to previous ones. Imbalance in data occurs when systematic societal, historical, or other ambient bias is present in the data. In this paper, we address both biases by proposing a hybrid fairness-aware recommender system. Our model provides efficient and accurate recommendations by incorporating multiple user-user and item-item similarity measures, content, and demographic information, while addressing recommendation biases. We implement our model using a powerful and expressive probabilistic programming language called probabilistic soft logic. We experimentally evaluate our approach on a popular movie recommendation dataset, showing that our proposed model can provide more accurate and fairer recommendations, compared to a state-of-the art fair recommender system.

In this paper, we study a novel task that learns to compose music from natural language. Given the lyrics as input, we propose a melody composition model that generates lyrics-conditional melody as well as the exact alignment between the generated melody and the given lyrics simultaneously. More specifically, we develop the melody composition model based on the sequence-to-sequence framework. It consists of two neural encoders to encode the current lyrics and the context melody respectively, and a hierarchical decoder to jointly produce musical notes and the corresponding alignment. Experimental results on lyrics-melody pairs of 18,451 pop songs demonstrate the effectiveness of our proposed methods. In addition, we apply a singing voice synthesizer software to synthesize the "singing" of the lyrics and melodies for human evaluation. Results indicate that our generated melodies are more melodious and tuneful compared with the baseline method.

By Snehal Shah In the movie Avengers: Age of Ultron, the Iron Man and Bruce Banner – a genius scientist when he wasn't playing Hulk – built an artificial intelligence system named'Ultron' to help protect the earth. But Ultron – the peacekeeping programme embedded in a synthetic body – turned hostile, making it his mission to eradicate humans from the face of the earth. As earth's fate hung in the balance, the mightiest of Avengers had to come together to save the planet from complete annihilation. Does this, another Marvel Comic story turned into a sci-fi Hollywood film, have a semblance of realism? A couple of years ago a unique experimental self-driving car was released on New Jersey roads, that was not coded or programmed by engineers.

A creepy AI transfers the facial expressions of one person to another to create eerily realistic'deep fake' videos. The software accurately flips a segment of one video - such as the mouth of a character - to the style of another to create life-like fake clips. A video produced by the team transferred the mouth movements of British comedian John Oliver onto the face of US talk-show host Stephen Colbert. Researchers warned the technology could be used to create fake news clips that falsely put words into the mouths of politicians or other powerful figures. An AI transfers the facial expressions of one person to another to create eerily realistic'deep fake' videos.

The iPhone X, a dramatically redesigned model released last fall, got rid of the home button and introduced facial-recognition technology to unlock the device. It was the first mass-market smartphone to demand a $1,000 starting price. Although the iPhone X didn't fulfill analysts' lofty sales expectations, it fared well enough for Apple to up the ante with the bigger model, whose price is expected to unveil Wednesday.

Deep learning is a subfield of machine learning and it comprises several approaches to tackling the single most important goal of AI research: allowing computers to model our world well enough to exhibit something like what we humans call intelligence. On a basic conceptual level, deep learning approaches share a very basic trait. DL algorithms interpret the raw data through multiple processing layers. Each of these layers takes the output of the previous one as its input and creates a more abstract representation of it. As a result, the more data is being fed into the right algorithm, the more general are the rules and features that it's able to infer in relation to a given scenario and, therefore, the apter it gets at handling new, similar situations.

When he plays a techno show, Moritz Simon Geist doesn't reach for a laptop. Instead, he calls on his army of sonic robots--a collection of small, motorized creations that click, clank, and whirr in an intricate mechanical symphony. Geist composes robotic electronic music, a burgeoning genre of electro jams that relies on hardware, not software, to engineer electronic sounds and beats. His forthcoming EP, The Material Turn, debuts in October with four tracks made entirely from self-fashioned instruments--futuristic robo-kalimbas, a droning guitar, and salvaged hard drives turned into percussive beat machines. Watching Geist play music is a little like watching a mad scientist in a lab.