Technology
Apple publishes recovery instructions for bug that crashes iOS devices | Technology
Hey Siri, can you help me get access to my texts again after that strange message I received consisting of non-Latin characters crashed my iPhone? In advance of a software update to fix the bug, yes it can. Apple has published a temporary workaround for iOS users affected by the bug, which was found to affect iPhones earlier in the week, and subsequently revealed to be capable of affecting the Apple Watch, iPads and Macs. In a new article published on its support website, Apple outlined a three-step process, enabling people to open their Messages app again: The booby-trapped text message presented characters including those from Arabic, Chinese and Marathi in a specific sequence that choked Apple's CoreText system, crashing the recipient's device and leaving them unable to access their messages. Related: iPhone text message bug can crash Apple Watch, iPad and Mac too Pranksters seized upon the message to crash friends' iPhones, but security experts have warned that this kind of bug has the potential to be used for more malicious purposes in the future.
NEWS | MOL Group Announces Freshhh 2015 Winners
MOL Group announced yesterday the winners of the Freshhh 2015 competition, which sees students from all over the world compete in technology and business strategy simulations related to the oil and gas industry. 'Just Ask Siri', consisting of three students from the Prague University of Economics and the Czech Technical University, was awarded first place, with Hungary's'Oil's Creed' and Slovenia's'Decore' teams placing in second and third respectively. All three teams will now be given the opportunity to join MOL Group's graduate recruitment and development program. MOL Group HR Vice President Zdravka Demeter Bubalo commented in a company statement: "We congratulate the top three teams for winning the Freshhh competition 2015. I would like to thank all participants for their endless efforts during the competition. It is incredible to see how young students work with such difficult real-life cases and always find new solutions. The outstanding results from the participants and number of applications are showing us once more that we are heading in the right direction in order to attract top talents of the oil and gas industry."
Spectral MLE: Top-$K$ Rank Aggregation from Pairwise Comparisons
This paper explores the preference-based top-$K$ rank aggregation problem. Suppose that a collection of items is repeatedly compared in pairs, and one wishes to recover a consistent ordering that emphasizes the top-$K$ ranked items, based on partially revealed preferences. We focus on the Bradley-Terry-Luce (BTL) model that postulates a set of latent preference scores underlying all items, where the odds of paired comparisons depend only on the relative scores of the items involved. We characterize the minimax limits on identifiability of top-$K$ ranked items, in the presence of random and non-adaptive sampling. Our results highlight a separation measure that quantifies the gap of preference scores between the $K^{\text{th}}$ and $(K+1)^{\text{th}}$ ranked items. The minimum sample complexity required for reliable top-$K$ ranking scales inversely with the separation measure irrespective of other preference distribution metrics. To approach this minimax limit, we propose a nearly linear-time ranking scheme, called \emph{Spectral MLE}, that returns the indices of the top-$K$ items in accordance to a careful score estimate. In a nutshell, Spectral MLE starts with an initial score estimate with minimal squared loss (obtained via a spectral method), and then successively refines each component with the assistance of coordinate-wise MLEs. Encouragingly, Spectral MLE allows perfect top-$K$ item identification under minimal sample complexity. The practical applicability of Spectral MLE is further corroborated by numerical experiments.
A trust-region method for stochastic variational inference with applications to streaming data
Theis, Lucas, Hoffman, Matthew D.
Stochastic variational inference allows for fast posterior inference in complex Bayesian models. However, the algorithm is prone to local optima which can make the quality of the posterior approximation sensitive to the choice of hyperparameters and initialization. We address this problem by replacing the natural gradient step of stochastic varitional inference with a trust-region update. We show that this leads to generally better results and reduced sensitivity to hyperparameters. We also describe a new strategy for variational inference on streaming data and show that here our trust-region method is crucial for getting good performance.
Learning Relational Event Models from Video
Dubba, Krishna S. R., Cohn, Anthony G., Hogg, David C., Bhatt, Mehul, Dylla, Frank
Event models obtained automatically from video can be used in applications ranging from abnormal event detection to content based video retrieval. When multiple agents are involved in the events, characterizing events naturally suggests encoding interactions as relations. Learning event models from this kind of relational spatio-temporal data using relational learning techniques such as Inductive Logic Programming (ILP) hold promise, but have not been successfully applied to very large datasets which result from video data. In this paper, we present a novel framework REMIND (Relational Event Model INDuction) for supervised relational learning of event models from large video datasets using ILP. Efficiency is achieved through the learning from interpretations setting and using a typing system that exploits the type hierarchy of objects in a domain. The use of types also helps prevent over generalization. Furthermore, we also present a type-refining operator and prove that it is optimal. The learned models can be used for recognizing events from previously unseen videos. We also present an extension to the framework by integrating an abduction step that improves the learning performance when there is noise in the input data. The experimental results on several hours of video data from two challenging real world domains (an airport domain and a physical action verbs domain) suggest that the techniques are suitable to real world scenarios.
The Ceteris Paribus Structure of Logics of Game Forms
Grossi, Davide, Lorini, Emiliano, Schwarzentruber, Francois
The article introduces a ceteris paribus modal logic, called CP, interpreted on the equivalence classes induced by finite sets of propositional atoms. This logic is studied and then used to embed three logics of strategic interaction, namely atemporal STIT, the coalition logic of propositional control (CL PC) and the starless fragment of the dynamic logic of propositional assignments (DL PA). The embeddings highlight a common ceteris paribus structure underpinning the key operators of all these apparently very different logics and show, we argue, remarkable similarities behind some of the most influential formalisms for reasoning about strategic interaction.
Coactive Learning
Shivaswamy, Pannaga, Joachims, Thorsten
We propose Coactive Learning as a model of interaction between a learning system and a human user, where both have the common goal of providing results of maximum utility to the user. Interactions in the Coactive Learning model take the following form: at each step, the system (e.g. search engine) receives a context (e.g. query) and predicts an object (e.g. ranking); the user responds by correcting the system if necessary, providing a slightly improved but not necessarily optimal object as feedback. We argue that such preference feedback can be inferred in large quantity from observable user behavior (e.g., clicks in web search), unlike the optimal feedback required in the expert model or the cardinal valuations required for bandit learning. Despite the relaxed requirements for the feedback, we show that it is possible to adapt many existing online learning algorithms to the coactive framework. In particular, we provide algorithms that achieve square root regret in terms of cardinal utility, even though the learning algorithm never observes cardinal utility values directly. We also provide an algorithm with logarithmic regret in the case of strongly convex loss functions. An extensive empirical study demonstrates the applicability of our model and algorithms on a movie recommendation task, as well as ranking for web search.
Compositional Vector Space Models for Knowledge Base Completion
Neelakantan, Arvind, Roth, Benjamin, McCallum, Andrew
Knowledge base (KB) completion adds new facts to a KB by making inferences from existing facts, for example by inferring with high likelihood nationality(X,Y) from bornIn(X,Y). Most previous methods infer simple one-hop relational synonyms like this, or use as evidence a multi-hop relational path treated as an atomic feature, like bornIn(X,Z) -> containedIn(Z,Y). This paper presents an approach that reasons about conjunctions of multi-hop relations non-atomically, composing the implications of a path using a recursive neural network (RNN) that takes as inputs vector embeddings of the binary relation in the path. Not only does this allow us to generalize to paths unseen at training time, but also, with a single high-capacity RNN, to predict new relation types not seen when the compositional model was trained (zero-shot learning). We assemble a new dataset of over 52M relational triples, and show that our method improves over a traditional classifier by 11%, and a method leveraging pre-trained embeddings by 7%.
A deep-structured fully-connected random field model for structured inference
Wong, Alexander, Shafiee, Mohammad Javad, Siva, Parthipan, Wang, Xiao Yu
There has been significant interest in the use of fully-connected graphical models and deep-structured graphical models for the purpose of structured inference. However, fully-connected and deep-structured graphical models have been largely explored independently, leaving the unification of these two concepts ripe for exploration. A fundamental challenge with unifying these two types of models is in dealing with computational complexity. In this study, we investigate the feasibility of unifying fully-connected and deep-structured models in a computationally tractable manner for the purpose of structured inference. To accomplish this, we introduce a deep-structured fully-connected random field (DFRF) model that integrates a series of intermediate sparse auto-encoding layers placed between state layers to significantly reduce computational complexity. The problem of image segmentation was used to illustrate the feasibility of using the DFRF for structured inference in a computationally tractable manner. Results in this study show that it is feasible to unify fully-connected and deep-structured models in a computationally tractable manner for solving structured inference problems such as image segmentation.
Microsoft to expand Cortana voice assistant to iOS and Android - CNET
Microsoft's Cortana voice assistant is coming to IOS and Android. In a blog post published Tuesday, Joe Belfiore, head of Microsoft's Operating Systems Group, announced the upcoming launch of a Cortana app for Android phones and iPhones that will work as a companion to Cortana on your Windows 10 PC. The new Phone companion app will allow you to install and use Cortana for Android and iOS as part of Microsoft's goal to connect your Windows PC to whichever smartphone you own. Will Cortana for Android and iOS have the same capabilities as the version on your Windows Phone? You won't be able to use Cortana to open apps and change settings in the Android or iOS editions.