Note: For those of you who prefer watching videos, please feel free to play the above video on the same content. While speech synthesis has come a long way since Kratzenstein's vowel organ that could produce the five vowel sounds, it is a whole'nother level of challenge to transform text to natural-sounding speech. Recent developments in deep learning have provided us a new approach to the challenge and in this article, we shall briefly introduce a mainstream text-to-speech method before the deep learning era, then explore models like WaveNet that Google's text-to-speech API service is now using for lifelike speech synthesis. If you pause and think for a moment about how you can perform text-to-speech, you would probably formulate a method that is very similar to the concatenative approach. In concatenative text-to-speech, texts are broken down into smaller units such as phonemes, and the corresponding recordings of the units are then combined to form a complete speech.

As the scope of artificial intelligence is growing, big tech organizations are investing, a huge amount of money, towards the research and development of AI solutions. These organizations are in a race to achieve AI supremacy. In this article, we will explore the Top 5 companies with market share, leading the arena of AI. The exponential growth of Amazon never ceases. The market Revenue of Amazon for the year 2019, accounted for US$2,130 million, with a market share of 7.1%.

I work at ADEO Group, the 3rd home-improvement retail company worldwide. I have been working for a few months in the data team, in helping some business teams become more data-driven and providing them data solutions, with the help of a great team of data scientists and analysts. My background is mainly business, and my playgrounds are customer services and customer relationship management businesses. And what I like to do in my free time is to play basketball, surf, to collect vinyl records. They all want to get actionable insights from the 50K textual feedback we get every month (web, mail, social, messaging, chatbots, marketing studies, UX surveys, employee surveys, web crawling,…).

So you are impressed with what you have heard so far about the latest and greatest deep learning and you would like to use it to solve your business problem. Well… the good news is you can -- the bad news is that it might not help you solve your problem. Let's take a step back and try to understand the relationship between artificial intelligence, machine learning, and deep learning: AI or Artificial Intelligence is a large universe of concepts that includes all ML or machine learning. In the same vein, ML encompasses all that is deep learning, or DL. Artificial intelligence: the intelligence displayed by machines that simulate human and animal intelligence.

There is a new challenge workload on the horizon, one where few can afford to compete. But for those who can, it will spark a rethink in what is possible from even the most powerful traditional supercomputers. It might sound odd that it can be collected under the banner of language modeling since that invokes speech and text analysis and generation. But emerging workloads and research show how far this is from traditional natural language processing. Over the next several years, language models will likely become far more general purpose, encompassing an unimaginable range of problem types. Being able to have a world described through language and rendered as an image or video, or even asking text-based questions about the world with answers based on a system's understanding of our nuanced reality sounds like science fiction.

The scale of systems employed in industrial environments demands a large number of sensors to facilitate meticulous monitoring and functioning. These requirements could potentially lead to inefficient system designs. The data coming from various sensors are often correlated due to the underlying relations in the system parameters that the sensors monitor. In theory, it should be possible to emulate the output of certain sensors based on other sensors. Tapping into such possibilities holds tremendous advantages in terms of reducing system design complexity. In order to identify the subset of sensors whose readings can be emulated, the sensors must be grouped into clusters. Complex systems generally have a large quantity of sensors that collect and store data over prolonged periods of time. This leads to the accumulation of massive amounts of data. In this paper we propose an end-to-end algorithmic solution, to realise virtual sensors in such systems. This algorithm splits the dataset into blocks and clusters each of them individually. It then fuses these clustering solutions to obtain a global solution using an Ant Colony inspired technique, FAC2T. Having grouped the sensors into clusters, we select representative sensors from each cluster. These sensors are retained in the system while the other sensors readings are emulated by applying supervised learning algorithms.

Deep energy renovation of building stock came more into focus in the European Union due to energy efficiency related directives. Many buildings that must undergo deep energy renovation are old and may lack design/renovation documentation, or possible degradation of materials might have occurred in building elements over time. Thermal transmittance (i.e. U-value) is one of the most important parameters for determining the transmission heat losses through building envelope elements. It depends on the thickness and thermal properties of all the materials that form a building element. In-situ U-value can be determined by ISO 9869-1 standard (Heat Flux Method - HFM). Still, measurement duration is one of the reasons why HFM is not widely used in field testing before the renovation design process commences. This paper analyzes the possibility of reducing the measurement time by conducting parallel measurements with one heat-flux sensor. This parallelization could be achieved by applying a specific class of the Artificial Neural Network (ANN) on HFM results to predict unknown heat flux based on collected interior and exterior air temperatures. After the satisfying prediction is achieved, HFM sensor can be relocated to another measuring location. Paper shows a comparison of four ANN cases applied to HFM results for a measurement held on one multi-layer wall - multilayer perceptron with three neurons in one hidden layer, long short-term memory with 100 units, gated recurrent unit with 100 units and combination of 50 long short-term memory units and 50 gated recurrent units. The analysis gave promising results in term of predicting the heat flux rate based on the two input temperatures. Additional analysis on another wall showed possible limitations of the method that serves as a direction for further research on this topic.

Natural Language Processing(NLP) has evolved significantly over the last decade. This paper highlights the most important milestones of this period, while trying to pinpoint the contribution of each individual model and algorithm to the overall progress. Furthermore, it focuses on issues still remaining to be solved, emphasizing on the groundbreaking proposals of Transformers, BERT, and all the similar attention-based models.

The right to be forgotten states that a data subject has the right to erase their data from an entity storing it. In the context of machine learning (ML), it requires the ML model provider to remove the data subject's data from the training set used to build the ML model, a process known as \textit{machine unlearning}. While straightforward and legitimate, retraining the ML model from scratch upon receiving unlearning requests incurs high computational overhead when the training set is large. To address this issue, a number of approximate algorithms have been proposed in the domain of image and text data, among which SISA is the state-of-the-art solution. It randomly partitions the training set into multiple shards and trains a constituent model for each shard. However, directly applying SISA to the graph data can severely damage the graph structural information, and thereby the resulting ML model utility. In this paper, we propose GraphEraser, a novel machine unlearning method tailored to graph data. Its contributions include two novel graph partition algorithms, and a learning-based aggregation method. We conduct extensive experiments on five real-world datasets to illustrate the unlearning efficiency and model utility of GraphEraser. We observe that GraphEraser achieves 2.06$\times$ (small dataset) to 35.94$\times$ (large dataset) unlearning time improvement compared to retraining from scratch. On the other hand, GraphEraser achieves up to $62.5\%$ higher F1 score than that of random partitioning. In addition, our proposed learning-based aggregation method achieves up to $112\%$ higher F1 score than that of the majority vote aggregation.

Neural predictive models have achieved groundbreaking performance improvements in various natural language processing tasks. However, most of neural predictive models suffer from the lack of explainability of predictions, limiting their practical utility, especially in the medical domain. This paper proposes a novel neural predictive framework coupled with large pre-trained language models to make a prediction and generate its corresponding explanation simultaneously. We conducted a preliminary empirical study on Chinese medical multiple-choice question answering, English natural language inference and commonsense question answering tasks. The experimental results show that the proposed approach can generate reasonable explanations for its predictions even with a small-scale training explanation text. The proposed method also achieves improved prediction accuracy on three datasets, which indicates that making predictions can benefit from generating the explanation in the decision process.