Whereas people learn many different types of knowledge from diverse experiences over many years, most current machine learning systems acquire just a single function or data model from just a single data set. We propose a never-ending learning paradigm for machine learning, to better reflect the more ambitious and encompassing type of learning performed by humans. As a case study, we describe the Never-Ending Language Learner (NELL), which achieves some of the desired properties of a never-ending learner, and we discuss lessons learned. NELL has been learning to read the web 24 hours/day since January 2010, and so far has acquired a knowledge base with over 80 million confidence-weighted beliefs (e.g., servedWith(tea, biscuits)). NELL has also learned millions of features and parameters that enable it to read these beliefs from the web. Additionally, it has learned to reason over these beliefs to infer new beliefs, and is able to extend its ontology by synthesizing new relational predicates. NELL can be tracked online at http://rtw.ml.cmu.edu, and followed on Twitter at @CMUNELL.

Mobile robots are often proposed as a favorable substitute to human correspondence in emergency response, disaster relief, and environmental monitoring scenarios. In this work, the next iteration of the Aquapod is proposed as a method to facilitate collection of subsurface liquid samples in order to assess toxicity levels in a body of water. This amphibious small form-factor robot is equipped with a buoyancy control unit, detachable fluidic sampling unit, and a wide range of sensing and processing capabilities. The robot was designed to move and collect water samples to a maximum depth of ten meters. Its unique form of tumbling locomotion results in a versatile platform that can be used in both terrestrial and aquatic environments leveraging its high mobility-to-size ratio.

As mobile robots decrease in size so does their ability to traverse rough terrain. New forms of locomotion beyond the basic wheel are being explored to overcome this fault. This paper expands on the mechanical design of a previous robot with a high mobility-to-size ratio. To accomplish high mobility the robot uses tumbling as its form of locomotion. By actively involving the body of the robot in the locomotion it can scale larger obstacles and will not get stuck in compliant terrain like similar sized wheeled robots. To accommodate real-world environments the new design has been waterproofed and moreover can be completely submerged in water to operate on a lake or stream floor. Additionally, this robot is equipped with a buoyancy control unit which will allow the robot to either sink or float in water, offering many unique applications in environmental monitoring and surveillance. This paper describes a first generation, radio controlled prototype of the design.

We consider here the problem of building a never-ending language learner; that is, an intelligent computer agent that runs forever and that each day must (1) extract, or read, information from the web to populate a growing structured knowledge base, and (2) learn to perform this task better than on the previous day. In particular, we propose an approach and a set of design principles for such an agent, describe a partial implementation of such a system that has already learned to extract a knowledge base containing over 242,000 beliefs with an estimated precision of 74% after running for 67 days, and discuss lessons learned from this preliminary attempt to build a never-ending learning agent.

We consider the problem of semi-supervised learning to extract categories (e.g., academic fields, athletes) and relations (e.g., PlaysSport(athlete, sport)) from web pages, starting with a handful of labeled training examples of each category or relation, plus hundreds of millions of unlabeled web documents. Semi-supervised training using only a few labeled examples is typically unreliable because the learning task is underconstrained. This paper pursues the thesis that much greater accuracy can be achieved by further constraining the learning task, by coupling the semi-supervised training of many extractors for different categories and relations. We characterize several ways in which the training of category and relation extractors can be coupled, and present experimental results demonstrating significantly improved accuracy as a result.

Process-induced variations and sub-threshold leakage in bulk-Si technology limit the scaling of SRAM into sub-32 nm nodes. New device architectures are being considered to improve$V_{T}$control and reduce short channel effects. Among the likely candidates, FinFETs are the most attractive option because of their good scalability and possibilities for further SRAM performance and yield enhancement through independent gating. The enhancements to read/write margins and yield are investigated in detail for two cell designs employing independently gated FinFETs. It is shown that FinFET-based 6-T SRAM cells designed with pass-gate feedback (PGFB) achieve significant improvements in the cell read stability without area penalty. The write-ability of the cell can be improved through the use of pull-up write gating (PUWG) with a separate write word line (WWL). The benefits of these two approaches are complementary and additive, allowing for simultaneous read and write yield enhancements when the PGFB and PUWG designs are used in combination.

We provide a theoretical analysis of the chance accuracies of large collections of classifiers. We show that on problems with small numbers of examples, some classifier can perform well by random chance, and we derive a theorem to explicitly calculate this accuracy. We use this theorem to provide a principled feature selection criterion for sparse, high-dimensional problems. We evaluate this method on microarray and fMRI datasets and show that it performs very close to the optimal accuracy obtained from an oracle. We also show that on the fMRI dataset this technique chooses relevant features successfully while another state-of-the-art method, the False Discovery Rate (FDR), completely fails at standard significance levels.

We study the problem of correcting spelling mistakes in text using memory-based learning techniques and a very large database of token n-gram occurrences in web text as training data. Our approach uses the context in which an error appears to select the most likely candidate from words which might have been intended in its place. Using a novel correction algorithm and a massive database of training data, we demonstrate higher accuracy on correcting real- word errors than previous work, and very high accuracy at a new task of ranking corrections to non-word errors given by a standard spelling correction package.