Despite their impressive performance, Deep Neural Networks (DNNs) typically underperform Gradient Boosting Trees (GBTs) on many tabular-dataset learning tasks. We propose that applying a different regularization coefficient to each weight might boost the performance of DNNs by allowing them to make more use of the more relevant inputs. However, this will lead to an intractable number of hyperparameters. Here, we introduce Regularization Learning Networks (RLNs), which overcome this challenge by introducing an efficient hyperparameter tuning scheme that minimizes a new Counterfactual Loss. Our results show that RLNs significantly improve DNNs on tabular datasets, and achieve comparable results to GBTs, with the best performance achieved with an ensemble that combines GBTs and RLNs. RLNs produce extremely sparse networks, eliminating up to 99.8% of the network edges and 82% of the input features, thus providing more interpretable models and reveal the importance that the network assigns to different inputs. RLNs could efficiently learn a single network in datasets that comprise both tabular and unstructured data, such as in the setting of medical imaging accompanied by electronic health records.

We introduce a conceptually simple and scalable framework for continual learning domains where tasks are learned sequentially. Our method is constant in the number of parameters and is designed to preserve performance on previously encountered tasks while accelerating learning progress on subsequent problems. This is achieved through training two neural networks: A knowledge base, capable of solving previously encountered problems, which is connected to an active column that is employed to efficiently learn the current task. After learning a new task, the active column is distilled into the knowledge base, taking care to protect any previously learnt tasks. This cycle of active learning (progression) followed by consolidation (compression) requires no architecture growth, no access to or storing of previous data or tasks, and no task-specific parameters. Thus, it is a learning process that may be sustained over a lifetime of tasks while supporting forward transfer and minimising forgetting. We demonstrate the progress & compress approach on sequential classification of handwritten alphabets as well as two reinforcement learning domains: Atari games and 3D maze navigation.

The space of graphs is characterized by a non-trivial geometry, which often complicates performing inference in practical applications. A common approach is to use embedding techniques to represent graphs as points in a conventional Euclidean space, but non-Euclidean spaces are often better suited for embedding graphs. Among these, constant curvature manifolds (CCMs), like hyperspheres and hyperboloids, offer a computationally tractable way to compute metric, yet non-Euclidean, geodesic distances. In this paper, we introduce a novel adversarial graph embedding technique to represent graphs on CCMs, and exploit such a mapping for detecting changes in stationarity in a graph-generating process. To this end, we introduce a novel family of change detection tests operating by means of distances on CCMs. We perform experiments on synthetic graph streams, and on sequences of functional networks extracted from iEEG data with the aim of detecting the onset of epileptic seizures. We show that our methods are able to detect extremely small changes in the graph-generating process, consistently outperforming solutions based on Euclidean embeddings. The general nature of our framework highlights its potential to be extended to other applications characterized by graph data or non-Euclidean geometries.

A non-vector-based dissimilarity measure is proposed by combining vector-based distance metrics and set operations. This proposed compound dissimilarity measure (CDM) is applicable to quantify similarity of collections of attribute/feature pairs where not all attributes are present in all collections. This is a typical challenge in the context of e.g., fingerprinting-based positioning (FbP). Compared to vector-based distance metrics (e.g., Minkowski), the merits of the proposed CDM are i) the data do not need to be converted to vectors of equal dimension, ii) shared and unshared attributes can be weighted differently within the assessment, and iii) additional degrees of freedom within the measure allow to adapt its properties to application needs in a data-driven way. We indicate the validity of the proposed CDM by demonstrating the improvements of the positioning performance of fingerprinting-based WLAN indoor positioning using four different datasets, three of them publicly available. When processing these datasets using CDM instead of conventional distance metrics the accuracy of identifying buildings and floors improves by about 5% on average. The 2d positioning errors in terms of root mean squared error (RMSE) are reduced by a factor of two, and the percentage of position solutions with less than 2m error improves by over 10%.

Dialogue act recognition is an important part of natural language understanding. We investigate the way dialogue act corpora are annotated and the learning approaches used so far. We find that the dialogue act is context-sensitive within the conversation for most of the classes. Nevertheless, previous models of dialogue act classification work on the utterance-level and only very few consider context. We propose a novel context-based learning method to classify dialogue acts using a character-level language model utterance representation, and we notice significant improvement. We evaluate this method on the Switchboard Dialogue Act corpus, and our results show that the consideration of the preceding utterances as a a context of the current utterance improves dialogue act detection.

To improve the quality of computation experience for mobile devices, mobile-edge computing (MEC) is a promising paradigm by providing computing capabilities in close proximity within a sliced radio access network (RAN), which supports both traditional communication and MEC services. Nevertheless, the design of computation offloading policies for a virtual MEC system remains challenging. Specifically, whether to execute a computation task at the mobile device or to offload it for MEC server execution should adapt to the time-varying network dynamics. In this paper, we consider MEC for a representative mobile user in an ultra-dense sliced RAN, where multiple base stations (BSs) are available to be selected for computation offloading. The problem of solving an optimal computation offloading policy is modelled as a Markov decision process, where our objective is to maximize the long-term utility performance whereby an offloading decision is made based on the task queue state, the energy queue state as well as the channel qualities between MU and BSs. To break the curse of high dimensionality in state space, we first propose a double deep Q-network (DQN) based strategic computation offloading algorithm to learn the optimal policy without knowing a priori knowledge of network dynamics. Then motivated by the additive structure of the utility function, a Q-function decomposition technique is combined with the double DQN, which leads to novel learning algorithm for the solving of stochastic computation offloading. Numerical experiments show that our proposed learning algorithms achieve a significant improvement in computation offloading performance compared with the baseline policies.

The consumer relationship with artificial intelligence (AI) has had its rough patches. But initial terror at the specter of job-gobbling automation and affront at the insurgence of'inhuman' interactions is waning as the promise of support, entertainment, connectivity and even protection emerges - presenting retailers who dare to deal in the un-real with major opportunities. The sentiment was starkly illustrated when Italian fashion giant Prada promoted its Autumn/Winter 2018 catwalk show in league with'Instagram's first virtual influencer', superseding the fashion industry's vast pool of human bloggers, many of whom now represent retail's heftiest marketing spends. Lil Miquela, who may or may not be based on an LA-based blogger, is a bone fide CGI superstar; as of May 2018, she has a whopping 1.1M Instagram followers, a figure that's rocketed from an already robust 600k during the February show. She teased followers with videos, GIFs of the new collection and archival pieces, Instagram stories (micro videos) and even a tour of the venue via a drone that she controlled before the show kicked off.

Callaway edited the video on his home computer, and then yanked out the hard drive and threw it into the ocean, the agent said. A third councilman arranged for a computer expert to help edit the video and blur the woman's face; charges against him were later dropped.

The news of the recent Cambridge Analytica scandal did not just bring privacy concerns to the fore, it also signified how valuable a commodity data has become in today's information age. It showed us how impactful the process of leveraging data can be, and why companies across the board are adopting Machine Learning to learn from their data and completely transform their businesses. Despite the clear need, global enterprises and start-ups are struggling to scale their Machine Learning initiatives. One of the major factors limiting scale is the inability to acquire and retain the right talent. In a recent Zinnov Talent Hotbeds Forecast, we analyzed multiple cities for Machine Learning talent and predicted new hubs that would emerge over the next two decades. In the study, we found that Global top 500 R&D spenders together employ over 92,000 employees skilled in Machine Learning technologies and that 32% of this number are employed by the world's Tech Giants such as Google, Amazon, Facebook, and Microsoft.

On May 25, a new law called the General Data Protection Regulation (GDPR) is going into effect in the European Union. The law was created to protect EU citizens from potential abuses, like the recent Cambridge Analytica scandal. Though the timing may seem coincidental, this law has been in the works for more than four years. GDPR will replace the Data Protection Directive (95/46/EC) of 1995. Under GDPR, companies can be fined up to 4% of their worldwide annual revenue from the previous financial year.