Reinforcement learning is effective in optimizing policies for recommender systems. Current solutions mostly focus on model-free approaches, which require frequent interactions with a real environment, and thus are expensive in model learning. Offline evaluation methods, such as importance sampling, can alleviate such limitations, but usually request a large amount of logged data and do not work well when the action space is large. In this work, we propose a model-based reinforcement learning solution which models the user-agent interaction for offline policy learning via a generative adversarial network. To reduce bias in the learnt policy, we use the discriminator to evaluate the quality of generated sequences and rescale the generated rewards. Our theoretical analysis and empirical evaluations demonstrate the effectiveness of our solution in identifying patterns from given offline data and learning policies based on the offline and generated data.

Knowledge distillation is a widely used technique for model compression. We posit that the teacher model used in a distillation setup, captures relationships between classes, that extend beyond the original dataset. We empirically show that a teacher model can transfer this knowledge to a student model even on an {\it out-of-distribution} dataset. Using this approach, we show promising results on MNIST, CIFAR-10, and Caltech-256 datasets using unlabeled image data from different sources. Our results are encouraging and help shed further light from the perspective of understanding knowledge distillation and utilizing unlabeled data to improve model quality.

Robustness to capitalization errors is a highly desirable characteristic of named entity recognizers, yet we find standard models for the task are surprisingly brittle to such noise. Existing methods to improve robustness to the noise completely discard given orthographic information, mwhich significantly degrades their performance on well-formed text. We propose a simple alternative approach based on data augmentation, which allows the model to \emph{learn} to utilize or ignore orthographic information depending on its usefulness in the context. It achieves competitive robustness to capitalization errors while making negligible compromise to its performance on well-formed text and significantly improving generalization power on noisy user-generated text. Our experiments clearly and consistently validate our claim across different types of machine learning models, languages, and dataset sizes.

Most of the existing question answering models can be largely compiled into two categories: i) open domain question answering models that answer generic questions and use large-scale knowledge base along with the targeted web-corpus retrieval and ii) closed domain question answering models that address focused questioning area and use complex deep learning models. Both the above models derive answers through textual comprehension methods. Due to their inability to capture the pedagogical meaning of textual content, these models are not appropriately suited to the educational field for pedagogy. In this paper, we propose an on-the-fly conceptual network model that incorporates educational semantics. The proposed model preserves correlations between conceptual entities by applying intelligent indexing algorithms on the concept network so as to improve answer generation. This model can be utilized for building interactive conversational agents for aiding classroom learning.

While machine-type communication (MTC) devices generate considerable amounts of data, they often cannot process the data due to limited energy and computation power. To empower MTC with intelligence, edge machine learning has been proposed. However, power allocation in this paradigm requires maximizing the learning performance instead of the communication throughput, for which the celebrated water-filling and max-min fairness algorithms become inefficient. To this end, this paper proposes learning centric power allocation (LCPA), which provides a new perspective to radio resource allocation in learning driven scenarios. By employing an empirical classification error model that is supported by learning theory, the LCPA is formulated as a nonconvex nonsmooth optimization problem, and is solved by majorization minimization (MM) framework. To get deeper insights into LCPA, asymptotic analysis shows that the transmit powers are inversely proportional to the channel gain, and scale exponentially with the learning parameters. This is in contrast to traditional power allocations where quality of wireless channels is the only consideration. Last but not least, to enable LCPA in large-scale settings, two optimization algorithms, termed mirror-prox LCPA and accelerated LCPA, are further proposed. Extensive numerical results demonstrate that the proposed LCPA algorithms outperform traditional power allocation algorithms, and the large-scale algorithms reduce the computation time by orders of magnitude compared with MM-based LCPA but still achieve competing learning performance.

A new paper from researchers at the University of Southern California's Brain and Creativity Institute considers a novel path toward creating robots with'feelings.' The key, according to researchers Kinson Man and Antonio Damasio, is homestasis, a self-preservation principle by which living creatures seek to maintain internal biological equilibrium by avoiding certain environments or kinds of stimuli. Were robots to be programmed with a homeostatic sense of self-preservation, would that put them on a path toward developing true feelings? According to a Science News report on the paper, Man and Damasio consider the most promising lead for feeling robots to come through the combination of soft robotics and deep learning, which when combined might approximate a homeostatic reaction to negative environmental stimuli. Man and Domasio point to a 1954 experiment by W. Ross Ashby that demonstrated how homeostatic sensing might be translated into robotics.

The machines are taking over the world! Well, maybe not quite yet, but AI and machine learning are already running much more of the world than you might realize. You could launch a career pioneering these tech innovations with the Machine Learning and Artificial Intelligence Certification Bundle. Today, you can sign up for only $29. Machine learning is the future, and careers are coming with it.

Think back to school and science class: You probably conducted an experiment where you placed an alarm clock (set to go off in 5 minutes) under a glass jar and the teacher pumped all the air out of the jar. When the alarm went off, you couldn't hear it, right? Without air, sound does not travel – nature abhors a vacuum. This is also true of artificial intelligence - it cannot survive in a vacuum and needs a rich ecosystem of data where it can thrive. This can only be achieved by integrating "trustworthy AI" systems with the rest of an organization's IT landscape.

Not many scientists get solicited for photo ops, but for Daphne Koller it's a regular occurrence. "It happens at pretty much any event that has tech people," Koller says when asked about one recent snapshot. It's not like I feel like this is something I deserve." Selfie requests are just one sign of Koller's stardom, earned from more than 20 years bridging computer science, biology and education. She chalked up a string of accolades along the way: getting a master's degree from Jerusalem's Hebrew University at 18; becoming a Stanford University professor focused on machine learning at 26; winning, nearly a decade later, a Mac Arthur "genius grant" for research that combined artificial intelligence and genomics; and cofounding $1 billion (valuation) Coursera, an early platform to let people around the world take university classes for free. The next act for this 51-year-old innovator: Insitro, a firm in South San Francisco that aims to find new drugs by sorting through masses of data. If it succeeds, it will have overturned how drugs get discovered. Lab biologists typically focus on a few specific proteins as drug targets. If those fail, data scientists make suggestions for others to try. Insitro, on the other hand, wants to collect much more data before the biologists go off on their hunt. It will leverage advances in bioengineering (such as Crispr gene editing) and in software that enables computers to see things that escape humans. Koller describes her aha moment this way: "Machine learning is now doing amazing things if you give it enough data.

The rise of technology within the education sector over the last few decades has been astounding. This is certainly the case if we consider that teaching with technology has become pervasive in almost every classroom environment. Within today's classroom, for example, we find ourselves surrounded by devices such as smart boards, AV, computers, laptops, tablets and phones, to name but a few technologies which are now being integrated into teaching. We have also seen the rise of the virtual learning environment and blended learning, alongside a significant rise in online education. This has allowed distance learning to take new forms and shapes and to reach greater audiences around the world.