Autonomous agents can learn by imitating teacher demonstrations of the intended behavior. Hierarchical control policies are ubiquitously useful for such learning, having the potential to break down structured tasks into simpler sub-tasks, thereby improving data efficiency and generalization. In this paper, we propose a variational inference method for imitation learning of a control policy represented by parametrized hierarchical procedures (PHP), a program-like structure in which procedures can invoke sub-procedures to perform sub-tasks. Our method discovers the hierarchical structure in a dataset of observation-action traces of teacher demonstrations, by learning an approximate posterior distribution over the latent sequence of procedure calls and terminations. Samples from this learned distribution then guide the training of the hierarchical control policy. We identify and demonstrate a novel benefit of variational inference in the context of hierarchical imitation learning: in decomposing the policy into simpler procedures, inference can leverage acausal information that is unused by other methods. Training PHP with variational inference outperforms LSTM baselines in terms of data efficiency and generalization, requiring less than half as much data to achieve a 24% error rate in executing the bubble sort algorithm, and to achieve no error in executing Karel programs.

Knowledge Distillation (KD) has made remarkable progress in the last few years and become a popular paradigm for model compression and knowledge transfer . However, almost all existing KD algorithms are data-driven, i.e., relying on a large amount of original training data or alternative data, which is usually unavailable in real-world scenarios. In this paper, we devote ourselves to this challenging problem and propose a novel adversarial distillation mechanism to craft a compact student model without any real-world data. W e introduce a model discrepancy to quantificationally measure the difference between student and teacher models and construct an optimizable upper bound. In our work, the student and the teacher jointly act the role of the discriminator to reduce this discrepancy, when a generator adversari-ally produces some "hard samples" to enlarge it. Extensive experiments demonstrate that the proposed data-free method yields comparable performance to existing data-driven methods. More strikingly, our approach can be directly extended to semantic segmentation, which is more complicated than classification, and our approach achieves state-of-the-art results. Code and pretrained models are available at https://github.com/VainF/

Policy distillation in deep reinforcement learning provides an effective way to transfer control policies from a larger network to a smaller untrained network without a significant degradation in performance. However, policy distillation is underexplored in deep reinforcement learning, and existing approaches are computationally inefficient, resulting in a long distillation time. In addition, the effectiveness of the distillation process is still limited to the model capacity. We propose a new distillation mechanism, called real-time policy distillation, in which training the teacher model and distilling the policy to the student model occur simultaneously. Accordingly, the teacher's latest policy is transferred to the student model in real time. This reduces the distillation time to half the original time or even less and also makes it possible for extremely small student models to learn skills at the expert level. We evaluated the proposed algorithm in the Atari 2600 domain. The results show that our approach can achieve full distillation in most games, even with compression ratios up to 1.7%.

Kenneth has had the privilege of living through several digital revolutions in his lifetime. From the mass adoption of computers in the early 90s to the advent of blockchain technology in the 2010s, he has developed a keen interest in the latest tech trends. Kenneth strongly believes that blockchain will have as much impact as the Internet and e-commerce combined.

"Don't worry, human intelligence will never be replaced by machines." That's what I was told as a freshman foreign languages student at a university. That was the time the concerns about the machine translation taking over the human, first came up. For an honest average playgoer, language is nothing but a set of words put in a specific order based on some (not so) simple rules. Learning languages is a grind.

There is a discrepancy in the quality of education worldwide, and it shouldn't come as a surprise that artificial intelligence (AI) education isn't an exception. While some students enjoy small classrooms with fancy electronic whiteboards and qualified instructors, others don't even have electricity. The world is digitally and economically divided, especially in the case of AI education. There are numerous barriers preventing students from acquiring AI skills besides a foundation in mathematics. Many underserved communities lack resources.

Laura Timms, Product Strategy Manager at MHR Analytics gives us a breakdown of the future effects of artificial intelligence on the human resources space. According to a recent survey, 82% of HR leaders believe their roles will be completely different in a decade's time. Big things are happening, with Artificial Intelligence (AI) taking a starring role. More than a third of the 500 companies we recently polled said they had adopted some form of AI in the past year, and almost half of the HR leaders we surveyed said that machine learning – a form of AI – will improve their HR function. AI is already being put to work in key areas such as recruitment, onboarding and employee development.

The recent increase in the availability of system condition monitoring data has lead to increases in the use of data-driven approaches for fault diagnostics. The accuracy of the fault detection and classification using these approaches is generally good when abundant labelled data on healthy and faulty system conditions exists and the diagnosis problem is formulated as a supervised learning task, i.e. supervised fault diagnosis. It is, however, relatively common in real situations that only a small fraction of the system condition monitoring data are labeled as healthy and the rest is unlabeled due to the uncertainty of the number and type of faults that may occur. In this case, supervised fault diagnosis performs poorly. Fault diagnosis with an unknown number and nature of faults is an open set learning problem where the knowledge of the faulty system is incomplete during training and the number and extent of the faults, of different types, can evolve during testing. In this paper, we propose to formulate the open set diagnostics problem as a semi-supervised learning problem and we demonstrate how it can be solved using a knowledge-induced learning approach with adaptive sampling variational autoencoders (KIL-AdaVAE) in combination with a one-class classifier. The fault detection and segmentation capability of the proposed method is demonstrated on a simulated case study using the Advanced Geared Turbofan 30000 (AGTF30) dynamical model under real flight conditions and induced faults of 17 fault types. The performance of the method is compared to the different learning strategies (supervised learning, supervised learning with embedding and semi-supervised learning) and deep learning algorithms. The results demonstrate that the proposed method is able to significantly outperform all other tested methods in terms of fault detection and fault segmentation.

Python and R are the de facto standard languages for data science, due to their ease of use and huge array of third party libraries for machine learning and analytics. This video provides an introduction to the ML Toolkit and demonstrates using InterSystems IRIS as both a standalone development platform and an orchestration tool for predictive modeling. Takeaway: The ML Toolkit enables machine learning and other complex application development in the R and Python languages.

Artificial intelligence (AI) has officially entered the higher education realm, both hypothetically and in early practice. According to the report Artificial Intelligence Market in the US Education Sector, AI will grow at a compound annual rate of 47.7 percent from 2018 to 2022. Several technological and educational powerhouses will contribute to that growth as they commit substantial resources and personnel to develop digital platforms that use AI. With all this buzz, there's understandably a good deal of confusion (and fear) around what AI is and what impact it might have on higher education. Defined as the development of computer systems that use reasoning, logic, and other human traits to perform tasks independently, AI offers ways for organizations to streamline processes and workflows through automation.