Iterative procedures for parameter estimation based on stochastic gradient descent allow the estimation to scale to massive data sets. However, in both theory and practice, they suffer from numerical instability. Moreover, they are statistically inefficient as estimators of the true parameter value. To address these two issues, we propose a new iterative procedure termed averaged implicit SGD (AI-SGD). For statistical efficiency, AI-SGD employs averaging of the iterates, which achieves the optimal Cram\'{e}r-Rao bound under strong convexity, i.e., it is an optimal unbiased estimator of the true parameter value. For numerical stability, AI-SGD employs an implicit update at each iteration, which is related to proximal operators in optimization. In practice, AI-SGD achieves competitive performance with other state-of-the-art procedures. Furthermore, it is more stable than averaging procedures that do not employ proximal updates, and is simple to implement as it requires fewer tunable hyperparameters than procedures that do employ proximal updates.

We propose using five data-driven community detection approaches from social networks to partition the label space for the task of multi-label classification as an alternative to random partitioning into equal subsets as performed by RAkELd: modularity-maximizing fastgreedy and leading eigenvector, infomap, walktrap and label propagation algorithms. We construct a label co-occurence graph (both weighted an unweighted versions) based on training data and perform community detection to partition the label set. We include Binary Relevance and Label Powerset classification methods for comparison. We use gini-index based Decision Trees as the base classifier. We compare educated approaches to label space divisions against random baselines on 12 benchmark data sets over five evaluation measures. We show that in almost all cases seven educated guess approaches are more likely to outperform RAkELd than otherwise in all measures, but Hamming Loss. We show that fastgreedy and walktrap community detection methods on weighted label co-occurence graphs are 85-92% more likely to yield better F1 scores than random partitioning. Infomap on the unweighted label co-occurence graphs is on average 90% of the times better than random paritioning in terms of Subset Accuracy and 89% when it comes to Jaccard similarity. Weighted fastgreedy is better on average than RAkELd when it comes to Hamming Loss.

Domain adaptation is the supervised learning setting in which the training and test data are sampled from different distributions: training data is sampled from a source domain, whilst test data is sampled from a target domain. This paper proposes and studies an approach, called feature-level domain adaptation (flda), that models the dependence between the two domains by means of a feature-level transfer model that is trained to describe the transfer from source to target domain. Subsequently, we train a domain-adapted classifier by minimizing the expected loss under the resulting transfer model. For linear classifiers and a large family of loss functions and transfer models, this expected loss can be comp uted or approximated analytically, and minimized efficiently. Our empirical evaluation of flda focuses on problems comprising binary and count data in which the transfer can be naturally modeled via a dropout distribution, which allows the classifier to adapt to differences in the marginal probability of features in the source and the target domain. Our experiments on several real-world problems show that flda performs on par with state-of-the-art domain-adaptation techniques. Keywords: Domain adaptation, transfer learning, sample selection bias, covariate shift, empirical risk minimization, dropout.

We introduce the Adaptive Skills, Adaptive Partitions (ASAP) framework that (1) learns skills (i.e., temporally extended actions or options) as well as (2) where to apply them. We believe that both (1) and (2) are necessary for a truly general skill learning framework, which is a key building block needed to scale up to lifelong learning agents. The ASAP framework can also solve related new tasks simply by adapting where it applies its existing learned skills. We prove that ASAP converges to a local optimum under natural conditions. Finally, our experimental results, which include a RoboCup domain, demonstrate the ability of ASAP to learn where to reuse skills as well as solve multiple tasks with considerably less experience than solving each task from scratch.

Machines are becoming more intelligent every year thanks to advances being made by companies like Google, Facebook, Microsoft, and many others. AI agents, as they're sometimes known, can already beat us at complex board games like Go and they're becoming more competent in a range of other areas. Now a London AI research lab owned by Google has carried out a study to make sure we can pull the plug on self-learning machines when we want to. DeepMind, acquired by Google for a reported 400 million in 2014, teamed up with scientists at the University of Oxford to find a way to make sure AI agents don't learn to prevent, or seek to prevent humans, from taking control. The peer-reviewed paper -- titled "Safely Interruptible Agents [PDF]" and published on the website of the Machine Intelligence Research Institute (MIRI) -- was written by Laurent Orseau, a research scientist at Google DeepMind, Stuart Armstrong at Oxford University's Future of Humanity Institute, and several others.

Three early-career engineers who are making a big difference in three very different areas of technology are to receive the Royal Academy's prestigious Silver Medal at the Academy Awards Dinner at the Tower of London on Thursday 23 June 2016. The Silver Medal celebrates outstanding personal contributions to UK engineering, which has resulted in successful market exploitation. Professor Dame Ann Dowling OM DBE FREng FRS, President of the Royal Academy of Engineering, says: "Damian Gardiner, Demis Hassabis and Tong Sun have all demonstrated the power of use-inspired research in taking ideas they have developed in academia and applying them to solve real-world problems. They are working with colleagues all over the world and making an enormous impact early in their careers that is both enriching academic knowledge and generating real economic benefit for the UK." Dr Damian Gardiner is taking the world of product authentication by storm, with his Cambridge University start-up company ilumink Limited acquired by Johnson Matthey's Process Technologies Division in 2015. They were keen to adopt his unique method of printing'liquid crystal' material onto any surface using an ink-jet printer.

The owner of a brand-new Tesla Model X SUV said the car suddenly accelerated at "maximum speed" by itself, jumped a curb and slammed into the side of a shopping mall while his wife was behind the wheel. The owner of the Model X, Puzant Ozbag, said the vehicle had been delivered only five days earlier to his home in Irvine, Calif., where the accident also took place. He said his wife had not activated any self-driving features at the time of the crash. "My wife is 45 years old and has had a driver's license almost 30 years and has a clean record. It's not like she's a 90-year-old person who's going to press the gas pedal instead of the brake," Ozbag said in an interview with Computerworld.

There is a lot of buzz in the media right now about new electric vehicle (EV) company Faraday Future (FF). The startup seemingly came out of nowhere and is currently making a huge impact on the EV world. Even with significant press and social media attention, FF is still secretive about its upcoming EV projects. Auto-blog giant, Jalopnik, published an article about just how "mysterious" the EV startup is. The company's intrigue stems not only from its uniquely designed concept car, the FFZERO1, but also from its leaders' forward thinking mentality regarding technological advancement – especially artificial intelligence (AI) – something that the automotive industry has barely tapped into.

Google employees, squeezed onto metal risers and standing in the back of a meeting room, erupted in cheers as newly arrived executive Andrew Conrad announced they would try to turn science fiction into reality: The tech giant had formed a biotech venture to create a futuristic device like Star Trek's iconic "Tricorder" diagnostic wizard -- and use it to cure cancer. Conrad, recalled an employee who was present, displayed images on the room's big screens showing nanoparticles tracking down cancer cells in the bloodstream and flashing signals to a Fitbit-style wristband. He promised a working prototype of the cancer early-detection device within six months. That was three years ago. Recently departed employees said the prototype didn't work as hoped, and the Tricorder project is floundering. Tricorder is not the only misfire for Google's ambitious and extravagantly funded biotech venture, now named Verily Life Sciences. It has announced three signature projects meant to transform medicine, and a STAT examination found that all of them are plagued by serious, if not fatal, scientific shortcomings, even as Verily has vigorously promoted their promise.

OLDWICK, N.J.--(BUSINESS WIRE)--In this A.M.BestTV episode, Neil Jacobstein, artificial intelligence and robotics co-chair, Singularity University, predicts a wave of change as web-based insurance competitors use intelligent automated systems to address risk in new ways. Click on http://www.ambest.com/v.asp?v jacobstein516 to view the entire program. "The insurance industry has been using artificial intelligence for over three decades," said Jacobstein. "They've used it in different forms, and it's gotten better and better. The underlying technology is now quite sophisticated but it's used actually throughout the industry." Jacobstein also believes that the usage of artificial intelligence is really something that will disrupt the industry over the next few years.