KSI offers free seminars and classes on trending IT topics. Please complete the form below if you wish to receive emails about our seminars and further information.

The Allen Institute for Artificial Intelligence (AI2) is a non-profit research institute in Seattle founded by Paul Allen and headed by Professor Oren Etzioni. The core mission of (AI2) is to contribute to humanity through high-impact AI research and engineering. We are actively seeking post docs and Research Scientists at all levels who are passionate about AI and who can help us achieve this core mission by teaming to construct AI systems with reasoning, learning and reading capabilities. AI2 Research Scientists will have a primary focus in one of these specific areas but will also have the opportunity to contribute and engage in a variety of other areas critical to our research and mission. These include opportunities to participate in or lead select R&D projects, work with management to develop the long term vision for knowledge systems R&D, take a leading role in overseeing and implementing software systems supporting AI2's research, author and present scientific papers and presentations for peer-reviewed journals and conferences, and help develop collaborative and strategic relationships with relevant academic, industrial, government, and standards organizations.

Automatic text summarization (ATS) has recently achieved impressive performance thanks to recent advances in deep learning and the availability of large-scale corpora. To make the summarization results more faithful, this paper presents an unsupervised approach that combines rhetorical structure theory, deep neural model and domain knowledge concern for ATS. This architecture mainly contains three components: domain knowledge base construction based on representation learning, attentional encoder-decoder model for rhetorical parsing and subroutine-based model for text summarization. Domain knowledge can be effectively used for unsupervised rhetorical parsing thus rhetorical structure trees for each document can be derived. In the unsupervised rhetorical parsing module, the idea of translation was adopted to alleviate the problem of data scarcity. The subroutine-based summarization model purely depends on the derived rhetorical structure trees and can generate content-balanced results. To evaluate the summary results without golden standard, we proposed an unsupervised evaluation metric, whose hyper-parameters were tuned by supervised learning. Experimental results show that, on a large-scale Chinese dataset, our proposed approach can obtain comparable performances compared with existing methods.

The multiple challenges of operating ever more complex environments are well known. The most common we hear when we are speaking with our customers and partners are based around the vast amount of data now being produced and the quality of it. These aren't new problems when it comes to network availability and performance monitoring. When I started working in this area in the late 1990s, Network Operations Centers (NOCs) were already drowning in the amount of data being produced. Back then, in the early days of systems and network management solutions, data would simply be discarded to avoid overloading the network management system.

In this paper we propose and study the novel problem of explaining node embeddings by finding embedded human interpretable subspaces in already trained unsupervised node representation embeddings. We use an external knowledge base that is organized as a taxonomy of human-understandable concepts over entities as a guide to identify subspaces in node embeddings learned from an entity graph derived from Wikipedia. We propose a method that given a concept finds a linear transformation to a subspace where the structure of the concept is retained. Our initial experiments show that we obtain low error in finding fine-grained concepts.

Causal inference is central to many areas of artificial intelligence, including complex reasoning, planning, knowledge-base construction, robotics, explanation, and fairness. An active community of researchers develops and enhances algorithms that learn causal models from data, and this work has produced a series of impressive technical advances. However, evaluation techniques for causal modeling algorithms have remained somewhat primitive, limiting what we can learn from experimental studies of algorithm performance, constraining the types of algorithms and model representations that researchers consider, and creating a gap between theory and practice. We argue for more frequent use of evaluation techniques that examine interventional measures rather than structural or observational measures, and that evaluate those measures on empirical data rather than synthetic data. We survey the current practice in evaluation and show that these are rarely used in practice. We show that such techniques are feasible and that data sets are available to conduct such evaluations. We also show that these techniques produce substantially different results than using structural measures and synthetic data.

John McCarthy, the father of Artificial Intelligence, describes it as "the science and engineering of making intelligent machines, especially intelligent computer programs". Artificial Intelligence is a way of making a computer and software related to computer which can think intelligently and autonomously, kind of similar to a human mind. In general understanding artificial intelligence is accomplished by studying how a human brain works while solving a problem and in what manner it learns and makes decisions, where outcomes of such kind of study are used as the basis of developing intelligent software and systems. Till now this field was dominated by quasi-artificial intelligent systems called "expert systems," which mainly used a rules-based decision-making process.1 In other words, we can interpret that these systems were not fully autonomous and, therefore, not truly intelligent, because they lacked the ability to learn and produce unpredictable results, and mostly they acted in a manner predetermined by their programming.2

We all know that feeling when the solution to a problem we have invested significant time into suddenly and unexpectedly reveals itself in a wondrous "a-ha" moment. Such an epiphany can be described as an enlightening realization that allows a problem or situation to be understood from a new and deeper perspective. Epiphanies, once deemed as insight from the divine, are relatively rare occurrences, but what if today's artificial intelligence (AI) tools can inspire and increase the frequency of epiphanies about the nature of very complex problems? BuildingIQ has set out to do exactly that --to move epiphanies out of the realm of the miraculous and into our everyday experience. We recently launched our powerful AI-driven inference engine, named Epiphany, which pulls together disparate data points within a given system; creates a virtualized network of that holistic system; and then learns how each point is connected and influenced by the other points in the network.

Knowledge-based biomedical data science (KBDS) involves the design and implementation of computer systems that act as if they knew about biomedicine. Such systems depend on formally represented knowledge in computer systems, often in the form of knowledge graphs. Here we survey the progress in the last year in systems that use formally represented knowledge to address data science problems in both clinical and biological domains, as well as on approaches for creating knowledge graphs. Major themes include the relationships between knowledge graphs and machine learning, the use of natural language processing, and the expansion of knowledge-based approaches to novel domains, such as Chinese Traditional Medicine and biodiversity.

What makes a task relatively more or less difficult for a machine compared to a human? Much AI/ML research has focused on expanding the range of tasks that machines can do, with a focus on whether machines can beat humans. Allowing for differences in scale, we can seek interesting (anomalous) pairs of tasks T, T'. We define interesting in this way: The "harder to learn" relation is reversed when comparing human intelligence (HI) to AI. While humans seems to be able to understand problems by formulating rules, ML using neural networks does not rely on constructing rules. We discuss a novel approach where the challenge is to "perform well under rules that have been created by human beings." We suggest that this provides a rigorous and precise pathway for understanding the difference between the two kinds of learning. Specifically, we suggest a large and extensible class of learning tasks, formulated as learning under rules. With these tasks, both the AI and HI will be studied with rigor and precision. The immediate goal is to find interesting groundtruth rule pairs. In the long term, the goal will be to understand, in a generalizable way, what distinguishes interesting pairs from ordinary pairs, and to define saliency behind interesting pairs. This may open new ways of thinking about AI, and provide unexpected insights into human learning.