Within machine learning, the supervised learning field aims at modeling the input-output relationship of a system, from past observations of its behavior. Decision trees characterize the input-output relationship through a series of nested $if-then-else$ questions, the testing nodes, leading to a set of predictions, the leaf nodes. Several of such trees are often combined together for state-of-the-art performance: random forest ensembles average the predictions of randomized decision trees trained independently in parallel, while tree boosting ensembles train decision trees sequentially to refine the predictions made by the previous ones. The emergence of new applications requires scalable supervised learning algorithms in terms of computational power and memory space with respect to the number of inputs, outputs, and observations without sacrificing accuracy. In this thesis, we identify three main areas where decision tree methods could be improved for which we provide and evaluate original algorithmic solutions: (i) learning over high dimensional output spaces, (ii) learning with large sample datasets and stringent memory constraints at prediction time and (iii) learning over high dimensional sparse input spaces.

Element Data, Inc., a decision support software platform that harnesses artificial intelligence and machine learning has acquired the technology assets and team of PV Cube, a Seattle area start-up. The acquisition expands the size of the team of existing engineers building the world's first cognitive decision engine. PV Cube's Co-Founders Vish Vadlamani and Phani Vaddadi, were most recently employed at Microsoft and led the development of Microsoft's knowledge fabric integrated within Cortana, Bing and other Microsoft products. Vadlamani and Vaddadi are named on a combined 45 awarded patents. Element Data's Chief Technology Officer Charles Davis said, "The community of sophisticated artificial intelligence and machine learning experts is in high demand. We are fortunate to have such highly regarded industry leaders on our team."

Our goal is to learn a semantic parser that maps natural language utterances into executable programs when only indirect supervision is available: examples are labeled with the correct execution result, but not the program itself. Consequently, we must search the space of programs for those that output the correct result, while not being misled by spurious programs: incorrect programs that coincidentally output the correct result. We connect two common learning paradigms, reinforcement learning (RL) and maximum marginal likelihood (MML), and then present a new learning algorithm that combines the strengths of both. The new algorithm guards against spurious programs by combining the systematic search traditionally employed in MML with the randomized exploration of RL, and by updating parameters such that probability is spread more evenly across consistent programs. We apply our learning algorithm to a new neural semantic parser and show significant gains over existing state-of-the-art results on a recent context-dependent semantic parsing task.

In the past years, a collection of hardware, software and online service have managed to bring changes and reforms to classrooms and teaching methods. But the true disruption of education is yet to arrive. Artificial Intelligence has proven its role as a game changing factor in an increasing number of fields, causing transformations unimaginable in the past. It's now showing glimmers of how it might forever change the learning process, one of the oldest skills that mankind has mastered. We've teamed up with Product Hunt to offer you the chance to win an all expense paid trip to TNW Conference 2017!

When an Icelander arrives at an office building and sees'Solarfri' posted, they need no further explanation for the empty premises: The word means'when staff get an unexpected afternoon off to enjoy good weather.' The people of this rugged North Atlantic island settled by Norsemen some 1,100 years ago have a unique dialect of Old Norse that has adapted to life at the edge of the Artic. Hundslappadrifa, for example, means'heavy snowfall with large flakes occurring in calm wind.' Linguistics experts wonder if this is the beginning of the end for the Icelandic tongue. Salome Sigurjonsdottir, 10, tests a voice-controlled television in an electronics store in Reykjavik. Linguistics experts, studying the future of a language spoken by fewer than 400,000 people in an increasingly globalized world, wonder if this is the beginning of the end for the Icelandic tongue.

When we offered CS231n (Deep Learning class) at Stanford, we intentionally designed the programming assignments to include explicit calculations involved in backpropagation on the lowest level. The students had to implement the forward and the backward pass of each layer in raw numpy. This is seemingly a perfectly sensible appeal - if you're never going to write backward passes once the class is over, why practice writing them? Are we just torturing the students for our own amusement? Some easy answers could make arguments along the lines of "it's worth knowing what's under the hood as an intellectual curiosity", or perhaps "you might want to improve on the core algorithm later", but there is a much stronger and practical argument, which I wanted to devote a whole post to: The problem with Backpropagation is that it is a leaky abstraction.

Informatica CIO Graeme Thompson looks at the rise of the machines and the role of the CIO in unlocking machine learning success. We live in an age characterised by the need to move fast, to learn from our mistakes, adapt and respond to change and predict future outcomes, so that we can gain competitive edge. Yet the pace at which business moves today and the availability of information means organisations are expected to make decisions and deliver change at a rate unheard of just a few years ago. We as humans don't have the processing power to accept all available inputs, apply the lessons we learn about what does and doesn't work quickly enough to drive incremental improvements, let alone big, disruptive innovations. Most current prediction models have their roots in statistics innovations dating back to the 19th and 20th centuries.

Disruptive technologies like Artificial Intelligence (AI) and Internet of Things (IoT) will enable organisations to deliver a million midday meals to school children in India, said a top official of software major Accenture on Thursday. "We have applied AI, IoT and block-chain to exponentially increase the number of midday meals served to children in state-run schools under'Million Meals' project with NGO Akshaya Patra," said Accenture Labs Managing Director Sanjay Podder in a statement here. Block-chain is a distributed database that compiles a growing list of ordered records called blocks. The US-based Accenture collaborated with the city-based Akshaya Patra Foundation, which operates the government-funded MidDay Meal Scheme in state-run and state-aided schools, in implementing the novel project with new IT products and solutions. "The project has also demonstrated how new technologies can help address challenges in mass meal production and delivery by revolutionising supply chain and operations," asserted Podder.

Teaching machines to accomplish tasks by conversing naturally with humans is challenging. Currently, developing task-oriented dialogue systems requires creating multiple components and typically this involves either a large amount of handcrafting, or acquiring costly labelled datasets to solve a statistical learning problem for each component. In this work we introduce a neural network-based text-in, text-out end-to-end trainable goal-oriented dialogue system along with a new way of collecting dialogue data based on a novel pipe-lined Wizard-of-Oz framework. This approach allows us to develop dialogue systems easily and without making too many assumptions about the task at hand. The results show that the model can converse with human subjects naturally whilst helping them to accomplish tasks in a restaurant search domain.

In recent years there is a growing interest in using deep representations for reinforcement learning. In this paper, we present a methodology and tools to analyze Deep Q-networks (DQNs) in a non-blind matter. Moreover, we propose a new model, the Semi Aggregated Markov Decision Process (SAMDP), and an algorithm that learns it automatically. The SAMDP model allows us to identify spatio-temporal abstractions directly from features and may be used as a sub-goal detector in future work. Using our tools we reveal that the features learned by DQNs aggregate the state space in a hierarchical fashion, explaining its success. Moreover, we are able to understand and describe the policies learned by DQNs for three different Atari2600 games and suggest ways to interpret, debug and optimize deep neural networks in reinforcement learning.