News Article | July 25, 2017
The Education Technology Industry Network, a division of The Software & Information Industry Association, released an important new report today: “Guidelines for Conducting and Reporting EdTech Impact Research in U.S. K-12 Schools.” Authored by Dr. Denis Newman and the research team at Empirical Education Inc., the Guidelines provide 16 best practice standards of research for publishers and developers of educational technologies. The Guidelines are a response to the changing research methods and policies driven by the accelerating pace of development and passage of the Every Student Succeeds Act (ESSA), which has challenged the static notion of evidence defined in NCLB. Recognizing the need for consensus among edtech providers, customers in the K-12 school market, and policy makers at all levels, SIIA is making these Guidelines freely available. “SIIA members recognize that changes in technology and policy have made evidence of impact an increasingly critical differentiator in the marketplace,” said Bridget Foster, senior VP and managing director of ETIN. “The Guidelines show how research can be conducted and reported within a short timeframe and still contribute to continuous product improvement.” “The Guidelines for research on edtech products is consistent with our approach to efficacy: that evidence of impact can lead to product improvement,” said Amar Kumar, senior vice president of Efficacy & Research at Pearson. “We appreciate ETIN’s leadership and Empirical Education’s efforts in putting together this clear presentation of how to use rigorous and relevant research to drive growth in the market.” The Guidelines draw on over a decade of experience in conducting research in the context of the U.S. Department of Education’s Institute of Education Sciences, and its Investing in Innovation program. “The current technology and policy environment provides an opportunity to transform how research is done,” said Dr. Newman, CEO of Empirical Education Inc. and lead author of the Guidelines. “Our goal in developing the new guidelines was to clarify current requirements in a way that will help edtech companies provide school districts with the evidence they need to consistently quantify the value of software tools. My thanks go to SIIA and the highly esteemed panel of reviewers whose contribution helped us provide the roadmap for the change that is needed.” “In light of the ESSA evidence standards and the larger movement toward evidence-based reform, publishers and software developers are increasingly being called upon to show evidence that their products make a difference with children,” said Guidelines peer reviewer Dr. Robert Slavin, director of the Center for Research and Reform in Education, Johns Hopkins University. “The ETIN Guidelines provide practical, sensible guidance to those who are ready to meet these demands.” ETIN’s goal is to improve the market for edtech products by advocating for greater transparency in reporting research findings. For that reason, it is actively working with government, policy organizations, foundations, and universities to gain the needed consensus for change. “As digital instructional materials flood the market place, state and local leaders need access to evidence-based research regarding the effectiveness of products and services. This guide is a great step in supporting both the public and private sector to help ensure students and teachers have access to the most effective resources for learning,” stated Christine Fox, Deputy Executive Director, SETDA. The Guidelines can be downloaded here: https://www.empiricaleducation.com/research-guidelines. About the Education Technology Industry Network (ETIN) and the Software & Information Industry Association (SIIA) ETIN is the leading voice for companies that provide software applications, digital content, online learning services and related technologies across the PK-20 sector. ETIN drives growth and innovation within the industry by providing leadership, advocacy, business development opportunities, government relations, and critical edtech market information. SIIA is an umbrella association representing 800+ technology, data, and media companies globally. For more information, visit siia.net/etin. About Empirical Education Inc. Empirical Education Inc. is a Silicon Valley-based research company that develops tools and services to provide the evidence K-12 school systems need to make evidence-based decisions about their programs, policies, and personnel. The company brings its research, data analysis, engineering, and project management expertise to customers including edtech companies and their investors, the U.S. Department of Education, foundations, leading research organizations, and state and local education agencies. Over the last decade, Empirical has worked with school systems to conduct dozens of rigorous experiments and now offers services to edtech companies for fast turn-around and low-cost impact studies of their products. For more information visit https://www.empiricaleducation.com.
Empirical | Date: 2017-01-03
A method for coupling a prosthesis to a spinal segment in a patient includes the steps of selecting first and second reference points disposed along the spinal segment and pre-operatively measuring a target distance. The target distance extends between the first and second reference points while the patient is in a preferred posture such as the standing position. A prosthesis is coupled to the spinal segment and the prosthesis is then intra-operatively adjusted in order to set the distance between the first and second reference points based on the target distance.
Agency: European Commission | Branch: H2020 | Program: RIA | Phase: INT-10-2015 | Award Amount: 2.49M | Year: 2016
Southeast Europe has seen a century of continuous transformation and transition the disappearance and emergence of states, political and legal systems, ideologies, institutions, and social classes. This has been accompanied by a stability of social practices resistant to change. Shaken by radically changing ideological and legal structures, citizens rely on customary and informal social networks of kin, symbolic kin, and friends for meeting economic needs, and on clan- or kin-related structures rather than the rule of law for security and protection. We trace the persistence of informal practices to: 1) the external origin of major transformations, including the transitions to and from socialism; 2) the incomplete character of change, which has tended to be replaced by equally radical but diametrically opposed projects; 3) the development of a buffer culture based on informal practices, directed to enabling people to survive under unstable conditions; and 4) the widening gap between formal institutions and informal social practices. The distance between proclaimed goals and existing practices represents the key challenge to the European integration of Balkan societies. The integration process could end with superficial change, behind which the real social life of corruption, clientelism, tension, inequality, and exclusion will continue to unfold. We propose to explicate the key formal and informal rules of the game, and to identify and decipher the unwritten rules which underpin tactical maneuvering between formal and informal institutions, in various spheres and at various levels of social life. These would then be compared to the demands and recommendations laid out in the key EU documents outlining expectations from Southeast European states. The goal is to contribute to the formulation of policy recommendations which would aim not to eradicate informal practices, but to close the gap between formal and informal institutions in Balkan societies.
Cherian A.,University of Minnesota |
Sra S.,Empirical |
Banerjee A.,University of Minnesota |
Papanikolopoulos N.,University of Minnesota
IEEE Transactions on Pattern Analysis and Machine Intelligence | Year: 2013
Covariance matrices have found success in several computer vision applications, including activity recognition, visual surveillance, and diffusion tensor imaging. This is because they provide an easy platform for fusing multiple features compactly. An important task in all of these applications is to compare two covariance matrices using a (dis)similarity function, for which the common choice is the Riemannian metric on the manifold inhabited by these matrices. As this Riemannian manifold is not flat, the dissimilarities should take into account the curvature of the manifold. As a result, such distance computations tend to slow down, especially when the matrix dimensions are large or gradients are required. Further, suitability of the metric to enable efficient nearest neighbor retrieval is an important requirement in the contemporary times of big data analytics. To alleviate these difficulties, this paper proposes a novel dissimilarity measure for covariances, the Jensen-Bregman LogDet Divergence (JBLD). This divergence enjoys several desirable theoretical properties and at the same time is computationally less demanding (compared to standard measures). Utilizing the fact that the square root of JBLD is a metric, we address the problem of efficient nearest neighbor retrieval on large covariance datasets via a metric tree data structure. To this end, we propose a K-Means clustering algorithm on JBLD. We demonstrate the superior performance of JBLD on covariance datasets from several computer vision applications. © 1979-2012 IEEE.
Hennig P.,Empirical |
Journal of Machine Learning Research | Year: 2013
Four decades after their invention, quasi-Newton methods are still state of the art in unconstrained numerical optimization. Although not usually interpreted thus, these are learning algorithms that fit a local quadratic approximation to the objective function. We show that many, including the most popular, quasi-Newton methods can be interpreted as approximations of Bayesian linear regression under varying prior assumptions. This new notion elucidates some shortcomings of classical algorithms, and lights the way to a novel nonparametric quasi-Newton method, which is able to make more efficient use of available information at computational cost similar to its predecessors. © 2013 Philipp Hennig and Martin Kiefel.
Hennig P.,Empirical |
Journal of Machine Learning Research | Year: 2012
Contemporary global optimization algorithms are based on local measures of utility, rather than a probability measure over location and value of the optimum. They thus attempt to collect low function values, not to learn about the optimum. The reason for the absence of probabilistic global optimizers is that the corresponding inference problem is intractable in several ways. This paper develops desiderata for probabilistic optimization algorithms, then presents a concrete algorithm which addresses each of the computational intractabilities with a sequence of approximations and explicitly addresses the decision problem of maximizing information gain from each evaluation. © 2012 Philipp Hennig and Christian J. Schuler.
Empirical | Date: 2016-02-25
A spinal implant for limiting flexion of the spine includes a tether structure for encircling adjacent spinal processes. Usually, a pair of compliance members will be provided as part of the tether structure for elastically limiting flexion while permitting an extension. A cross-member is provided between the compliance member or other portions of the tether structure to stabilize the tether structure and prevent misalignment after implantation.
Empirical | Date: 2015-12-28
The present invention provides systems and methods for deploying implantable devices within the body. The delivery and deployment systems include at least one catheter or an assembly of catheters for selectively positioning the lumens of the implant to within target vessels. Various deployment and attachment mechanisms are provided for selectively deploying the implants.
Empirical | Date: 2016-09-28
A system for restricting flexion of a spinal segment in a patient comprises a constraint device having a tether structure and a compliance member coupled with the tether structure. The tether structure is adapted to be coupled with a superior spinous process and a sacrum. The system also includes an anchor member that is anchored to the sacrum. The anchor member has an attachment feature that is adapted to couple with the constraint device.
Empirical | Date: 2016-02-25
A spinal treatment system includes a constraint device having an upper tether portion, a lower tether portion and a compliance member coupled therebetween. The upper tether portion is coupled with a superior spinous process of a spinal segment in a patient and the lower tether portion is coupled with an inferior spinous process or sacrum of the spinal segment. The length or tension in the constraint device is adjustable so that the construct of the tether portions and the compliance member provides a force resistant to flexion of the spinal segment. The system also includes a first prosthesis coupled with the spinal segment, wherein the constraint device modulates loads borne by the prosthesis or by tissue adjacent thereto.