Artificial Intelligence (AI) has recently shown its capabilities for almost every field of life. Machine Learning, which is a subset of AI, is a `HOT' topic for researchers. Machine Learning outperforms other classical forecasting techniques in almost all-natural applications. It is a crucial part of modern research. As per this statement, Modern Machine Learning algorithms are hungry for big data. Due to the small datasets, the researchers may not prefer to use Machine Learning algorithms. To tackle this issue, the main purpose of this survey is to illustrate, demonstrate related studies for significance of a semi-parametric Machine Learning framework called Grey Machine Learning (GML). This kind of framework is capable of handling large datasets as well as small datasets for time series forecasting likely outcomes. This survey presents a comprehensive overview of the existing semi-parametric machine learning techniques for time series forecasting. In this paper, a primer survey on the GML framework is provided for researchers. To allow an in-depth understanding for the readers, a brief description of Machine Learning, as well as various forms of conventional grey forecasting models are discussed. Moreover, a brief description on the importance of GML framework is presented.

Typical high quality text-to-speech (TTS) systems today use a two-stage architecture, with a spectrum model stage that generates spectral frames and a vocoder stage that generates the actual audio. High-quality spectrum models usually incorporate the encoder-decoder architecture with self-attention or bi-directional long short-term (BLSTM) units. While these models can produce high quality speech, they often incur O($L$) increase in both latency and real-time factor (RTF) with respect to input length $L$. In other words, longer inputs leads to longer delay and slower synthesis speed, limiting its use in real-time applications. In this paper, we propose a multi-rate attention architecture that breaks the latency and RTF bottlenecks by computing a compact representation during encoding and recurrently generating the attention vector in a streaming manner during decoding. The proposed architecture achieves high audio quality (MOS of 4.31 compared to groundtruth 4.48), low latency, and low RTF at the same time. Meanwhile, both latency and RTF of the proposed system stay constant regardless of input lengths, making it ideal for real-time applications.

Many learning tasks require us to deal with graph data which contains rich relational information among elements, leading increasing graph neural network (GNN) models to be deployed in industrial products for improving the quality of service. However, they also raise challenges to model authentication. It is necessary to protect the ownership of the GNN models, which motivates us to present a watermarking method to GNN models in this paper. In the proposed method, an Erdos-Renyi (ER) random graph with random node feature vectors and labels is randomly generated as a trigger to train the GNN to be protected together with the normal samples. During model training, the secret watermark is embedded into the label predictions of the ER graph nodes. During model verification, by activating a marked GNN with the trigger ER graph, the watermark can be reconstructed from the output to verify the ownership. Since the ER graph was randomly generated, by feeding it to a non-marked GNN, the label predictions of the graph nodes are random, resulting in a low false alarm rate (of the proposed work). Experimental results have also shown that, the performance of a marked GNN on its original task will not be impaired. Moreover, it is robust against model compression and fine-tuning, which has shown the superiority and applicability.

Algorithmic fairness has emphasized the role of biased data in automated decision outcomes. Recently, there has been a shift in attention to sources of bias that implicate fairness in other stages in the ML pipeline. We contend that one source of such bias, human preferences in model selection, remains under-explored in terms of its role in disparate impact across demographic groups. Using a deep learning model trained on real-world medical imaging data, we verify our claim empirically and argue that choice of metric for model comparison can significantly bias model-selection outcomes. While ML promised to remove human biases from decision making, the past several years have made it increasingly clear that automation is not a panacea with respect to fairer decision outcomes.

Non-negative Matrix Factorization (NMF) methods offer an appealing unsupervised learning method for real-time analysis of streaming spectral data in time-sensitive data collection, such as $\textit{in situ}$ characterization of materials. However, canonical NMF methods are optimized to reconstruct a full dataset as closely as possible, with no underlying requirement that the reconstruction produces components or weights representative of the true physical processes. In this work, we demonstrate how constraining NMF weights or components, provided as known or assumed priors, can provide significant improvement in revealing true underlying phenomena. We present a PyTorch based method for efficiently applying constrained NMF and demonstrate this on several synthetic examples. When applied to streaming experimentally measured spectral data, an expert researcher-in-the-loop can provide and dynamically adjust the constraints. This set of interactive priors to the NMF model can, for example, contain known or identified independent components, as well as functional expectations about the mixing of components. We demonstrate this application on measured X-ray diffraction and pair distribution function data from $\textit{in situ}$ beamline experiments. Details of the method are described, and general guidance provided to employ constrained NMF in extraction of critical information and insights during $\textit{in situ}$ and high-throughput experiments.

Model-free or learning-based control, in particular, reinforcement learning (RL), is expected to be applied for complex robotic tasks. Traditional RL requires a policy to be optimized is state-dependent, that means, the policy is a kind of feedback (FB) controllers. Due to the necessity of correct state observation in such a FB controller, it is sensitive to sensing failures. To alleviate this drawback of the FB controllers, feedback error learning integrates one of them with a feedforward (FF) controller. RL can be improved by dealing with the FB/FF policies, but to the best of our knowledge, a methodology for learning them in a unified manner has not been developed. In this paper, we propose a new optimization problem for optimizing both the FB/FF policies simultaneously. Inspired by control as inference, the optimization problem considers minimization/maximization of divergences between trajectory, predicted by the composed policy and a stochastic dynamics model, and optimal/non-optimal trajectories. By approximating the stochastic dynamics model using variational method, we naturally derive a regularization between the FB/FF policies. In numerical simulations and a robot experiment, we verified that the proposed method can stably optimize the composed policy even with the different learning law from the traditional RL. In addition, we demonstrated that the FF policy is robust to the sensing failures and can hold the optimal motion. Attached video is also uploaded on youtube: https://youtu.be/zLL4uXIRmrE

Predictive business process monitoring focuses on predicting future characteristics of a running process using event logs. The foresight into process execution promises great potentials for efficient operations, better resource management, and effective customer services. Deep learning-based approaches have been widely adopted in process mining to address the limitations of classical algorithms for solving multiple problems, especially the next event and remaining-time prediction tasks. Nevertheless, designing a deep neural architecture that performs competitively across various tasks is challenging as existing methods fail to capture long-range dependencies in the input sequences and perform poorly for lengthy process traces. In this paper, we propose ProcessTransformer, an approach for learning high-level representations from event logs with an attention-based network. Our model incorporates long-range memory and relies on a self-attention mechanism to establish dependencies between a multitude of event sequences and corresponding outputs. We evaluate the applicability of our technique on nine real event logs. We demonstrate that the transformer-based model outperforms several baselines of prior techniques by obtaining on average above 80% accuracy for the task of predicting the next activity. Our method also perform competitively, compared to baselines, for the tasks of predicting event time and remaining time of a running case.

Learning continually from non-stationary data streams is a long-standing goal and a challenging problem in machine learning. Recently, we have witnessed a renewed and fast-growing interest in continual learning, especially within the deep learning community. However, algorithmic solutions are often difficult to re-implement, evaluate and port across different settings, where even results on standard benchmarks are hard to reproduce. In this work, we propose Avalanche, an open-source end-to-end library for continual learning research based on PyTorch. Avalanche is designed to provide a shared and collaborative codebase for fast prototyping, training, and reproducible evaluation Figure 1: Operational representation of Avalanche with its of continual learning algorithms.

Human behavior refers to the way humans act and interact in response to a stimulus, internal or external. Understanding human behavior through observational study is one of the core methodologies in fields such as psychology and sociology (Margolin, Oliver, Gordis, O'hearn, Medina, Ghosh and Morland, 1998). Human behaviors encompass rich information: from emotional expression, processing, and regulation to the intricate dynamics of interactions, including the context and knowledge of interlocutors and their thinking and problem-solving intent (Li, Baucom and Georgiou, 2020). Furthermore, the behavioral constructs of interest are often dependent on the domain of interaction (Narayanan and Georgiou, 2013). Hence characterization of human behavior usually requires domain-specific knowledge and adequate windows of observation. Notably, across psychological health science and practice (Bone, Lee, Chaspari, Gibson and Narayanan, 2017) such as couple therapy (Christensen, Atkins, Berns, Wheeler, Baucom and Simpson, 2004), suicide cognition evaluation (Bryan, Rudd, Wertenberger, Etienne, Ray-Sannerud, Morrow, Peterson and Young-McCaughon, 2014) and addiction counseling (Xiao, Imel, Georgiou, Atkins and Narayanan, 2015), this is exemplified in the definition and derivation of a variety of domain-specific behavior constructs (e.g., blame and affect patterns exhibited by partners, suicidal ideation of an individual at risk, and empathy expressed by a therapist in the respective aforementioned domains) to support specific subsequent plan of action. Human speech offers rich information about the mental state and traits of the talkers. Vocal cues, including speech and spoken language as well as nonverbal vocalizations and disfluency patterns, have been shown to be informationally relevant in the context of human behavior (e.g., in marital interaction (Baucom, Atkins, Simpson and Christensen, 2009), in motivational interviewing (Amrhein, Miller, Yahne, Palmer and Fulcher, 2003; Imel, Barco, Brown, Baucom, Baer, Kircher and Atkins, 2014; Miller, Benefield and Tonigan, 1993)). Many automatic computational approaches that support measurement, analysis, and modeling of human behaviors from speech have been investigated in affective computing (Lee and Narayanan, 2005), social signal processing (Vinciarelli, Pantic and Bourlard, 2009) and behavioral signal processing (BSP) (Narayanan and Georgiou, 2013).

Replay is the reactivation of one or more neural patterns, which are similar to the activation patterns experienced during past waking experiences. Replay was first observed in biological neural networks during sleep, and it is now thought to play a critical role in memory formation, retrieval, and consolidation. Replay-like mechanisms have been incorporated into deep artificial neural networks that learn over time to avoid catastrophic forgetting of previous knowledge. Replay algorithms have been successfully used in a wide range of deep learning methods within supervised, unsupervised, and reinforcement learning paradigms. In this paper, we provide the first comprehensive comparison between replay in the mammalian brain and replay in artificial neural networks. We identify multiple aspects of biological replay that are missing in deep learning systems and hypothesize how they could be utilized to improve artificial neural networks.