Social media is awash with people talking about a new piece of software called GPT-3, which has been developed by OpenAI, an Elon Musk-backed artificial intelligence lab in San Francisco. GPT-3 (Generative Pre-training) is a language-generation tool capable of producing human-like text on demand. The software learned how to produce text by analyzing vast quantities on the internet and observing which letters and words tend to follow one another. OpenAI started releasing it to a select few people last week who had requested access to a private early version, and many of them have been blown away. "It's far more coherent than any AI language system I've ever tried," wrote entrepreneur Arram Sabeti in a blog post after testing. "All you have to do is write a prompt and it'll add text it thinks would plausibly follow.

In July, San Francisco-based OpenAI stunned everyone with the capabilities of its AI language model, the GPT-3. The text generator can pen fiction, compose poetry and write business memos — all without any human intervention. Simply put, it is being seen as a tool that brings machines a wee bit closer to mimicking human intelligence. A look at how AI has come this far.The GPT-3, or Generative Pre-training Transformer 3, can use half a sentence as input and type out the rest correctly. It doesn't stop at that. The text predictor can then type out a whole para, or a book, that make logical sense. The GPT-3, however, lacks the ability to reason abstractly. It is at a loss when faced with new ideas. ​First, what is the big fuss really?

At age 19, I read an interview of physicist Lee Smolin. The idea that a foreseeable limit exists on our understanding of physics by the end of my life was profoundly unsettling. I felt frustrated that I might never witness solutions to the great mysteries of science, no matter how hard I work. But… perhaps one can find a way to tear down this limit. Artificial intelligence presents a new regime of scientific inquiry, where we can automate the research process itself.

The implementation of robust, stable, and user-centered data analytics and machine learning models is confronted by numerous challenges in production and manufacturing. Therefore, a systematic approach is required to develop, evaluate, and deploy such models. The data-driven knowledge discovery framework provides an orderly partition of the data-mining processes to ensure the practical implementation of data analytics and machine learning models. However, the practical application of robust industry-specific data-driven knowledge discovery models faces multiple data-- and model-development--related issues. These issues should be carefully addressed by allowing a flexible, customized, and industry-specific knowledge discovery framework; in our case, this takes the form of the cross-industry standard process for data mining (CRISP-DM). This framework is designed to ensure active cooperation between different phases to adequately address data- and model-related issues. In this paper, we review several extensions of CRISP-DM models and various data-robustness-- and model-robustness--related problems in machine learning, which currently lacks proper cooperation between data experts and business experts because of the limitations of data-driven knowledge discovery models.

Corresponding author: Email address: changhaibin@pku.edu.cn Key Points: Two categories of innovative deep-learning based inverse modeling methods are proposed and compared. The deep-learning surrogate-based inversion methods can accelerate the inversion process significantly. Abstract Deep-learning has achieved good performance and shown great potential for solving forward and inverse problems. In this work, two categories of innovative deep-learning based inverse modeling methods are proposed and compared. The first category is deep-learning surrogate-based inversion methods, in which the Theory-guided Neural Network (TgNN) is constructed as a deep-learning surrogate for problems with uncertain model parameters. By incorporating physical laws and other constraints, the TgNN surrogate can be constructed with limited simulation runs and accelerate the inversion process significantly. Three TgNN surrogate-based inversion methods are proposed, including the gradient method, the iterative ensemble smoother (IES), and the training method. In TgNN-geo, two neural networks are introduced to approximate the respective random model parameters and the solution. Since the prior geostatistical information can be incorporated, the direct-inversion method based on TgNN-geo works well, even in cases with sparse spatial measurements or imprecise prior statistics. Although the proposed deep-learning based inverse modeling methods are general in nature, and thus applicable to a wide variety of problems, they are tested with several subsurface flow problems. It is found that satisfactory results are obtained with a high efficiency.

A central capability of a long-lived reinforcement learning (RL) agent is to incrementally adapt its behavior as its environment changes, and to incrementally build upon previous experiences to facilitate future learning in real-world scenarios. In this paper, we propose LifeLong Incremental Reinforcement Learning (LLIRL), a new incremental algorithm for efficient lifelong adaptation to dynamic environments. We develop and maintain a library that contains an infinite mixture of parameterized environment models, which is equivalent to clustering environment parameters in a latent space. The prior distribution over the mixture is formulated as a Chinese restaurant process (CRP), which incrementally instantiates new environment models without any external information to signal environmental changes in advance. During lifelong learning, we employ the expectation maximization (EM) algorithm with online Bayesian inference to update the mixture in a fully incremental manner. In EM, the E-step involves estimating the posterior expectation of environment-to-cluster assignments, while the M-step updates the environment parameters for future learning. This method allows for all environment models to be adapted as necessary, with new models instantiated for environmental changes and old models retrieved when previously seen environments are encountered again. Experiments demonstrate that LLIRL outperforms relevant existing methods, and enables effective incremental adaptation to various dynamic environments for lifelong learning.

Communication has become increasingly dynamic with the popularization of social networks and applications that allow people to express themselves and communicate instantly. In this scenario, distributed representation models have their quality impacted by new words that appear frequently or that are derived from spelling errors. These words that are unknown by the models, known as out-of-vocabulary (OOV) words, need to be properly handled to not degrade the quality of the natural language processing (NLP) applications, which depend on the appropriate vector representation of the texts. To better understand this problem and finding the best techniques to handle OOV words, in this study, we present a comprehensive performance evaluation of deep learning models for representing OOV words. We performed an intrinsic evaluation using a benchmark dataset and an extrinsic evaluation using different NLP tasks: text categorization, named entity recognition, and part-of-speech tagging. Although the results indicated that the best technique for handling OOV words is different for each task, Comick, a deep learning method that infers the embedding based on the context and the morphological structure of the OOV word, obtained promising results.

Transformers-based models, such as BERT, have been one of the most successful deep learning models for NLP. Unfortunately, one of their core limitations is the quadratic dependency (mainly in terms of memory) on the sequence length due to their full attention mechanism. To remedy this, we propose, BigBird, a sparse attention mechanism that reduces this quadratic dependency to linear. We show that BigBird is a universal approximator of sequence functions and is Turing complete, thereby preserving these properties of the quadratic, full attention model. Along the way, our theoretical analysis reveals some of the benefits of having $O(1)$ global tokens (such as CLS), that attend to the entire sequence as part of the sparse attention mechanism. The proposed sparse attention can handle sequences of length up to 8x of what was previously possible using similar hardware. As a consequence of the capability to handle longer context, BigBird drastically improves performance on various NLP tasks such as question answering and summarization. We also propose novel applications to genomics data.

This article provides an overview of Supervised Machine Learning (SML) with a focus on applications to banking. The SML techniques covered include Bagging (Random Forest or RF), Boosting (Gradient Boosting Machine or GBM) and Neural Networks (NNs). We begin with an introduction to ML tasks and techniques. This is followed by a description of: i) tree-based ensemble algorithms including Bagging with RF and Boosting with GBMs, ii) Feedforward NNs, iii) a discussion of hyper-parameter optimization techniques, and iv) machine learning interpretability. The paper concludes with a comparison of the features of different ML algorithms. Examples taken from credit risk modeling in banking are used throughout the paper to illustrate the techniques and interpret the results of the algorithms.

In recent years, we have witnessed the rise of deep learning. Deep neural networks have proved their success in many areas. However, the optimization of these networks has become more difficult as neural networks going deeper and datasets becoming bigger. Therefore, more advanced optimization algorithms have been proposed over the past years. In this study, widely used optimization algorithms for deep learning are examined in detail. To this end, these algorithms called adaptive gradient methods are implemented for both supervised and unsupervised tasks. The behaviour of the algorithms during training and results on four image datasets, namely, MNIST, CIFAR-10, Kaggle Flowers and Labeled Faces in the Wild are compared by pointing out their differences against basic optimization algorithms.