Search

We explore simple methods for adapting a trained multi-task UNet which predicts canopy cover and height to a new geographic setting using remotely sensed data without the need of training a domain-adaptive classifier and extensive fine-tuning. Extending previous research, we followed a selective alignment process to identify similar images in the two geographical domains and then tested an array of data-based unsupervised domain adaptation approaches in a zero-shot setting as well as with a small amount of fine-tuning. We find that the selective aligned data-based image matching methods produce promising results in a zero-shot setting, and even more so with a small amount of fine-tuning. These methods outperform both an untransformed baseline and a popular data-based image-to-image translation model. The best performing methods were pixel distribution adaptation and fourier domain adaptation on the canopy cover and height tasks respectively., Comment: ICLR 2024 Machine Learning for Remote Sensing (ML4RS) Workshop

There is currently considerable excitement within government about the potential of artificial intelligence to improve public service productivity through the automation of complex but repetitive bureaucratic tasks, freeing up the time of skilled staff. Here, we explore the size of this opportunity, by mapping out the scale of citizen-facing bureaucratic decision-making procedures within UK central government, and measuring their potential for AI-driven automation. We estimate that UK central government conducts approximately one billion citizen-facing transactions per year in the provision of around 400 services, of which approximately 143 million are complex repetitive transactions. We estimate that 84% of these complex transactions are highly automatable, representing a huge potential opportunity: saving even an average of just one minute per complex transaction would save the equivalent of approximately 1,200 person-years of work every year. We also develop a model to estimate the volume of transactions a government service undertakes, providing a way for government to avoid conducting time consuming transaction volume measurements. Finally, we find that there is high turnover in the types of services government provide, meaning that automation efforts should focus on general procedures rather than services themselves which are likely to evolve over time. Overall, our work presents a novel perspective on the structure and functioning of modern government, and how it might evolve in the age of artificial intelligence.

The field of machine learning has recently made significant progress in reducing the requirements for labelled training data when building new models. These `cheaper' learning techniques hold significant potential for the social sciences, where development of large labelled training datasets is often a significant practical impediment to the use of machine learning for analytical tasks. In this article we review three `cheap' techniques that have developed in recent years: weak supervision, transfer learning and prompt engineering. For the latter, we also review the particular case of zero-shot prompting of large language models. For each technique we provide a guide of how it works and demonstrate its application across six different realistic social science applications (two different tasks paired with three different dataset makeups). We show good performance for all techniques, and in particular we demonstrate how prompting of large language models can achieve high accuracy at very low cost. Our results are accompanied by a code repository to make it easy for others to duplicate our work and use it in their own research. Overall, our article is intended to stimulate further uptake of these techniques in the social sciences., Comment: 39 pages, 10 figures, 6 tables

Generative AI has the potential to transform how public services are delivered by enhancing productivity and reducing time spent on bureaucracy. Furthermore, unlike other types of artificial intelligence, it is a technology that has quickly become widely available for bottom-up adoption: essentially anyone can decide to make use of it in their day to day work. But to what extent is generative AI already in use in the public sector? Our survey of 938 public service professionals within the UK (covering education, health, social work and emergency services) seeks to answer this question. We find that use of generative AI systems is already widespread: 45% of respondents were aware of generative AI usage within their area of work, while 22% actively use a generative AI system. Public sector professionals were positive about both current use of the technology and its potential to enhance their efficiency and reduce bureaucratic workload in the future. For example, those working in the NHS thought that time spent on bureaucracy could drop from 50% to 30% if generative AI was properly exploited, an equivalent of one day per week (an enormous potential impact). Our survey also found a high amount of trust (61%) around generative AI outputs, and a low fear of replacement (16%). While respondents were optimistic overall, areas of concern included feeling like the UK is missing out on opportunities to use AI to improve public services (76%), and only a minority of respondents (32%) felt like there was clear guidance on generative AI usage in their workplaces. In other words, it is clear that generative AI is already transforming the public sector, but uptake is happening in a disorganised fashion without clear guidelines. The UK's public sector urgently needs to develop more systematic methods for taking advantage of the technology.

Allocation of scarce resources is a recurring challenge for the public sector: something that emerges in areas as diverse as healthcare, disaster recovery, and social welfare. The complexity of these policy domains and the need for meeting multiple and sometimes conflicting criteria has led to increased focus on the use of algorithms in this type of decision. However, little engagement between researchers across these domains has happened, meaning a lack of understanding of common problems and techniques for approaching them. Here, we performed a cross disciplinary literature review to understand approaches taken for different areas of algorithmic allocation including healthcare, organ transplantation, homelessness, disaster relief, and welfare. We initially identified 1070 papers by searching the literature, then six researchers went through them in two phases of screening resulting in 176 and 75 relevant papers respectively. We then analyzed the 75 papers from the lenses of optimization goals, techniques, interpretability, flexibility, bias, ethical considerations, and performance. We categorized approaches into human-oriented versus resource-oriented perspective, and individual versus aggregate and identified that 76% of the papers approached the problem from a human perspective and 60% from an aggregate level using optimization techniques. We found considerable potential for performance gains, with optimization techniques often decreasing waiting times and increasing success rate by as much as 50%. However, there was a lack of attention to responsible innovation: only around one third of the papers considered ethical issues in choosing the optimization goals while just a very few of them paid attention to the bias issues. Our work can serve as a guide for policy makers and researchers wanting to use an algorithm for addressing a resource allocation problem.

Existing research suggests that some transition activities (e.g., visits to kindergarten classrooms, home visits by teachers, sharing information with families) initiated by early care and education (ECE) settings, kindergarten teachers and elementary schools, and/or families can have small, positive associations with children's outcomes in kindergarten. This study examined combinations of kindergarten transition activities reported by both kindergarten teachers and parents to gain a comprehensive picture of supports that families experience. The authors identified: (1) Four common combinations of transition activities provided to and experienced by families, called Transition Activity Groups; (2) Whether these distinct combinations of transition activities differed for children who previously attended Head Start compared to other children; and (3) If those combinations of transition activities were associated with child and family outcomes and engagement in kindergarten. [This report was produced as part of the Head Start to Kindergarten (HS2K) Transitions Project. It was produced with the National P-3 Center.]

This article explores how the 'rules in use' from Ostrom's Institutional Analysis and Development Framework (IAD) can be developed as a context analysis approach for AI. AI risk assessment frameworks increasingly highlight the need to understand existing contexts. However, these approaches do not frequently connect with established institutional analysis scholarship. We outline a novel direction illustrated through a high-level example to understand how clinical oversight is potentially impacted by AI. Much current thinking regarding oversight for AI revolves around the idea of decision makers being in-the-loop and, thus, having capacity to intervene to prevent harm. However, our analysis finds that oversight is complex, frequently made by teams of professionals and relies upon explanation to elicit information. Professional bodies and liability also function as institutions of polycentric oversight. These are all impacted by the challenge of oversight of AI systems. The approach outlined has potential utility as a policy tool of context analysis aligned with the 'Govern and Map' functions of the National Institute of Standards and Technology (NIST) AI Risk Management Framework; however, further empirical research is needed. Our analysis illustrates the benefit of existing institutional analysis approaches in foregrounding team structures within oversight and, thus, in conceptions of 'human in the loop'., Comment: 19 pages

Calls for new metrics, technical standards and governance mechanisms to guide the adoption of Artificial Intelligence (AI) in institutions and public administration are now commonplace. Yet, most research and policy efforts aimed at understanding the implications of adopting AI tend to prioritize only a handful of ideas; they do not fully connect all the different perspectives and topics that are potentially relevant. In this position paper, we contend that this omission stems, in part, from what we call the relational problem in socio-technical discourse: fundamental ontological issues have not yet been settled--including semantic ambiguity, a lack of clear relations between concepts and differing standard terminologies. This contributes to the persistence of disparate modes of reasoning to assess institutional AI systems, and the prevalence of conceptual isolation in the fields that study them including ML, human factors, social science and policy. After developing this critique, we offer a way forward by proposing a simple policy and research design tool in the form of a conceptual framework to organize terms across fields--consisting of three horizontal domains for grouping relevant concepts and related methods: Operational, Epistemic, and Normative. We first situate this framework against the backdrop of recent socio-technical discourse at two premier academic venues, AIES and FAccT, before illustrating how developing suitable metrics, standards, and mechanisms can be aided by operationalizing relevant concepts in each of these domains. Finally, we outline outstanding questions for developing this relational approach to institutional AI research and adoption., Comment: 23 pages, 1 figure

Information on urban tree canopies is fundamental to mitigating climate change [1] as well as improving quality of life [2]. Urban tree planting initiatives face a lack of up-to-date data about the horizontal and vertical dimensions of the tree canopy in cities. We present a pipeline that utilizes LiDAR data as ground-truth and then trains a multi-task machine learning model to generate reliable estimates of tree cover and canopy height in urban areas using multi-source multi-spectral satellite imagery for the case study of Chicago., Comment: 4 pages, 4 figures, Submitted to Tackling Climate Change with Machine Learning: workshop at NeurIPS 2022

In 2013-14, Chicago Public Schools (CPS), the nation's third-largest school district, implemented several changes to the policies governing its pre-kindergarten (pre-k) application and enrollment process. Two policy changes that were central to the school district's strategy focused on increasing access to full-day pre-k for high-priority student groups: (1) increasing the overall number of full-day pre-k classrooms within school buildings; and (2) intentionally placing those full-day pre-k classrooms in neighborhoods with a large proportion of age-eligible, high-priority children and historically low rates of enrollment in CPS pre-k. This research examines whether and how access--defined as the distance to the closest school with full-day school-based pre-k and the number of full-day pre-k classrooms near potential students' homes--was related to pre-k enrollment and ongoing learning outcomes in early elementary school. In other words: Does the geographic placement of full-day pre-k classrooms within a school district matter? In prior work, it was found that after these policy changes in Chicago, there were more equitable patterns of pre-k access and enrollment. This brief presents new findings: Geographic placement of full-day pre-k matters not only for more equitable pre-k access and enrollment, but also for academic outcomes in early elementary school. Students--especially in high-priority student groups--who were eligible to enroll in pre-k after the policy changes had higher math scores and grades in early elementary school than did students eligible for pre-k before the policy changes. These benefits are partly explained by the fact that Black students and those living in lowest-income neighborhoods lived closer to full-day school-based pre-k options post-policy, which in turn was related to increased enrollment in full-day pre-k, higher kindergarten entry skills, and ultimately better 2nd grade academic outcomes.

We identify natural symmetries of each rigid higher braided category. Specifically, we construct a functorial action by the continuous group $\Omega \mathsf{O}(n)$ on each $\mathcal{E}_{n-1}$-monoidal $(g,d)$-category $\mathcal{R}$ in which each object is dualizable (for $n\geq 2$, $d \geq 0$, $d \leq g \leq \infty$). This action determines a canonical action by the continuous group $\Omega \mathbb{R}\mathbb{P}^{n-1}$ on the moduli space of objects of each such $\mathcal{R}$. In cases where the parameters $n$, $d$, and $g$ are small, we compare these continuous symmetries to known symmetries, which manifest as categorical identities., Comment: 20 pages, preliminary version, comments and suggested citations are welcome

The process and results of defining neighborhoods may look different depending on the purpose. In this brief, the authors focused on how neighborhoods in Chicago are defined. The initial goal of this work was to answer questions about how access to and enrollment in school-based pre-kindergarten (pre-k) may have varied by neighborhood characteristics. In conducting this work, the authors used a data-driven method for characterizing neighborhoods that leveraged publicly available census data and allowed for consideration of many neighborhood characteristics simultaneously. This method resulted in a parsimonious set of five neighborhood groupings in Chicago that simplified how the relationship of neighborhood characteristics to a host of educational and other outcomes was understood. This brief shares the approach to defining neighborhood groupings and their characteristics and the findings about them. First, the authors review prior neighborhood research and ways of characterizing Chicago community areas and neighborhoods, outlining how the current work builds on these previous efforts. Next the data-driven method for grouping neighborhoods is presented and the neighborhood groupings are described. Finally, the potential uses and implications of the work is discussed. The hope is that others in Chicago (or elsewhere) find this approach to defining neighborhood groupings useful -- both for conducting research studies and providing services or resources.

As pre-kindergarten (pre-k) expands across the country, school districts are making choices about where to place pre-k classrooms and developing policies for how families can apply and which children are enrolled. In doing so, districts are pulling policy levers that influence students' access to pre-k. Research shows that some families have less access to pre-k than others, which contributes to inequitable enrollment within districts. This descriptive study explores whether and how Chicago's school-based pre-k system was more equitable after the district implemented a set of policies focused on changing access to and enrollment in school-based pre-k. Specifically, we compare patterns of enrollment and geographic access (i.e., distance from home to a school with pre-k and number of pre-k classrooms nearby), and use multilevel logistic regression analyses to examine how access was related to enrollment for different student groups before and after the policy changes. Results show that both access to, and enrollment in, full-day pre-k expanded substantially among Black students, lowest-income students, and students living in mostly-Black neighborhoods, even as overall school-based pre-k enrollment remained relatively constant. There was a modest association between access to and enrollment in full-day pre-k (but not any pre-k), and this relationship became stronger following the policy changes, especially for high-priority student groups. Findings demonstrate that the geographic distribution of school-based pre-k (particularly full-day) may be an important policy lever for addressing inequities in student access and enrollment in Chicago, and have implications for other school districts implementing similar efforts nationwide. [For the research snapshot, see ED612486.]

We identify the group of framed diffeomorphisms of the torus as a semi-direct product of the torus with the braid group on 3 strands; we also identify the topological monoid of framed local-diffeomorphisms of the torus in similar terms. It follows that the framed mapping class group is this braid group. We show that the group of framed diffeomorphisms of the torus acts on twice-iterated Hochschild homology, and explain how this recovers a host of familiar symmetries. In the case of Cartesian monoidal structures, we show that this action extends to the monoid of framed local-diffeomorphisms of the torus. Based on this, we propose a definition of an unstable secondary cyclotomic structure, and show that iterated Hochschild homology possesses such in the Cartesian monoidal setting., Comment: 50 pages. Two appendices, one that collects a few facts about continuous monoids

We construct a circle-invariant trace from the factorization homology of the circle $ {\sf trace} \colon \int^\alpha_{{\mathbb S}^1} \\underline{\sf End}(V) \longrightarrow \uno $ associated to a dualizable object $V\in {\boldsymbol{\mathfrak X}}$ in a symmetric monoidal $\infty$-category. This proves a conjecture of To\"en--Vezzosi on existence of circle-invariant traces. Underlying our construction is a calculation of the factorization homology over the circle of the walking adjunction in terms of the paracyclic category of Getzler--Jones: $ \int_{{\mathbb S}^1} {\sf Adj} ~\simeq~ {\bDelta_{\circlearrowleft}}^{\triangleleft\!\triangleright} ~. $ This calculation exhibits a form of Poincar\'e duality for 1-dimensional factorization homology., Comment: 30 pages

We use differentiable programming and gradient descent to find unitary matrices that can be used in the period finding algorithm to extract period information from the state of a quantum computer post application of the oracle. The standard procedure is to use the inverse quantum Fourier transform. Our findings suggest that that this is not the only unitary matrix appropriate for the period finding algorithm, There exist several unitary matrices that can affect out the same transformation and they are significantly different from each other as well. These unitary matrices can be learned by an algorithm. Neural networks can be applied to differentiate such unitary matrices from randomly generated ones indicating that these unitaries do have characteristic features that cannot otherwise be discerned easily., Comment: 10 pages, 10 figures

The Pancreatic beta cell is an important target in diabetes research. For scalable modeling of beta cell ultrastructure, we investigate automatic segmentation of whole cell imaging data acquired through soft X-ray tomography. During the course of the study, both complete and partial ultrastructure annotations were produced manually for different subsets of the data. To more effectively use existing annotations, we propose a method that enables the application of partially labeled data for full label segmentation. For experimental validation, we apply our method to train a convolutional neural network with a set of 12 fully annotated data and 12 partially annotated data and show promising improvement over standard training that uses fully annotated data alone.

This chapter amalgamates some foundational developments and calculations in factorization homology., Comment: 50 pages, written for the new handbook of homotopy theory

We introduce \emph{flagged $(\infty,n)$-categories} and prove that they are equivalent to Segal sheaves on Joyal's category ${\mathbf\Theta}_n$. As such, flagged $(\infty,n)$-categories provide a model-independent formulation of Segal sheaves. This result generalizes the statement that $n$-groupoid objects in spaces are effective, as we explain and contextualize. Along the way, we establish a useful expression for the univalent-completion of such a Segal sheaf. Finally, we conjecture a characterization of flagged $(\infty,n)$-categories as stacks on $(\infty,n)$-categories that satisfy descent with respect to colimit diagrams that do not generate invertible $i$-morphisms for any $i$., Comment: 30 pages

For an arbitrary symmetric monoidal $\infty$-category $\mathcal{V}$, we define the factorization homology of $\mathcal{V}$-enriched $(\infty,1)$-categories over (possibly stratified) 1-manifolds and study some of its basic properties. In the case of spectral enrichment, we show that the value of factorization homology on a circle is topological Hochschild homology.

Assuming a conjecture about factorization homology with adjoints, we prove the cobordism hypothesis, after Baez-Dolan, Costello, Hopkins-Lurie, and Lurie., Comment: 36 pages, slightly revised; preliminary version, comments welcome

We construct a flagged $\infty$-category ${\sf Corr}$ of $\infty$-categories and bimodules among them. We prove that ${\sf Corr}$ classifies exponentiable fibrations. This representability of exponentiable fibrations extends that established by Lurie of both coCartesian fibrations and Cartesian fibrations, as they are classified by the $\infty$-category of $\infty$-categories and its opposite, respectively. We introduce the flagged $\infty$-subcategories ${\sf LCorr}$ and ${\sf RCorr}$ of ${\sf Corr}$, whose morphisms are those bimodules which are \emph{left final} and \emph{right initial}, respectively. We identify the notions of fibrations these flagged $\infty$-subcategories classify, and show that these $\infty$-categories carry universal left/right fibrations., Comment: 89 pages, accepted by Higher Structures. Differs slightly from published version: Lemma 3.10 and Proposition 3.11 have been corrected

We construct a pairing, which we call factorization homology, between framed manifolds and higher categories. The essential geometric notion is that of a vari-framing of a stratified manifold, which is a framing on each stratum together with a coherent system of compatibilities of framings along links between strata. Our main result constructs labeling systems on disk-stratified vari-framed $n$-manifolds from $(\infty,n)$-categories. These $(\infty,n)$-categories, in contrast with the literature to date, are not required to have adjoints. This allows the following conceptual definition: the factorization homology \[ \int_M\mathcal{C} \] of a framed $n$-manifold $M$ with coefficients in an $(\infty,n)$-category $\mathcal{C}$ is the classifying space of $\cC$-labeled disk-stratifications over $M$., Comment: 94 pages, 14 figures; contains an erratum which corrects the statement of the main theorem, based on the non-contractibility of vari-framed automorphisms of the hemispherical disk in higher-dimensions; differs from the published version

We show that conically smooth stratified spaces embed fully faithfully into $\infty$-categories. This articulates a stratified generalization of the homotopy hypothesis proposed by Grothendieck. As such, each $\infty$-category defines a stack on conically smooth stratified spaces, and we identify the descent conditions it satisfies. These include $\mathbb{R}^1$-invariance and descent for open covers and blow-ups, analogous to sheaves for the h-topology in $\mathbb{A}^1$-homotopy theory. In this way, we identify $\infty$-categories as striation sheaves, which are those sheaves on conically smooth stratified spaces satisfying the indicated descent. We use this identification to construct by hand two remarkable examples of $\infty$-categories: $\mathcal{B}{\sf un}$, an $\infty$-category classifying constructible bundles; and $\mathcal{E}{\sf xit}$, the absolute exit-path $\infty$-category. These constructions are deeply premised on stratified geometry, the key geometric input being a characterization of conically smooth stratified maps between cones and the existence of pullbacks for constructible bundles., Comment: 84 pages; numerous corrections made; definition of constructible bundle modified to add a condition on links, ensuring existence of pullbacks (Lemma 6.1.11)

We formulate a theory of pointed manifolds, accommodating both embeddings and Pontryagin-Thom collapse maps, so as to present a common generalization of Poincar\'e duality in topology and Koszul duality in $\mathcal{E}_n$-algebra., Comment: 58 pages, accepted by J. Inst. Math. Jussieu

We prove a duality for factorization homology which generalizes both usual Poincar\'e duality for manifolds and Koszul duality for $\mathcal{E}_n$-algebras. The duality has application to the Hochschild homology of associative algebras and enveloping algebras of Lie algebras. We interpret our result at the level of topological quantum field theory., Comment: 70 pages, final version, accepted by Communications in Mathematical Physics

This work forms a foundational study of factorization homology, or topological chiral homology, at the generality of stratified spaces with tangential structures. Examples of such factorization homology theories include intersection homology, compactly supported stratified mapping spaces, and Hochschild homology with coefficients. Our main theorem characterizes factorization homology theories by a generalization of the Eilenberg--Steenrod axioms; it can also be viewed as an analogue of the Baez--Dolan cobordism hypothesis formulated for the observables, rather than state spaces, of a topological quantum field theory. Using these axioms, we extend the nonabelian Poincar\'e duality of Salvatore and Lurie to the setting of stratified spaces -- this is a nonabelian version of the Poincar\'e duality given by intersection homology. We pay special attention to the simple case of singular manifolds whose singularity datum is a properly embedded submanifold and give a further simplified algebraic characterization of these homology theories. In the case of 3-manifolds with 1-dimensional submanifolds, these structure gives rise to knot and link homology theories., Comment: 52 pages

We develop a theory of conically smooth stratified spaces and their smooth moduli, including a notion of classifying maps for tangential structures. We characterize continuous space-valued sheaves on these conically smooth stratified spaces in terms of tangential data, and we similarly characterize 1-excisive invariants of stratified spaces. These results are based on the existence of open handlebody decompositions for conically smooth stratified spaces, an inverse function theorem, a tubular neighborhood theorem, an isotopy extension theorem, and functorial resolutions of singularities to smooth manifolds with corners., Comment: 92 pages, 5 figures; varies slightly from the published version

In this brief postscript to our paper "Integral transforms and Drinfeld centers in derived algebraic geometry", we describe a Morita equivalence for derived, categorified matrix algebras implied by theory developed since its appearance. We work in the setting of perfect stacks X and their stable infinity-categories Q(X) of quasicoherent sheaves. Perfect stacks include all varieties and common stacks in characteristic zero, and their stable infinity-categories of sheaves are well behaved refinements of their quasicoherent derived categories, satisfying natural analogues of common properties of function spaces. To a morphism of perfect stacks pi:X-->Y, we associate the categorified matrix algebra Q(X x_Y X) of sheaves on the derived fiber product equipped with its monoidal convolution product. We show that for pi faithfully flat (as a corollary of the 1-affineness theorem of Gaisgory) or for pi proper and surjective and X,Y smooth (as an application of proper descent, cf. Gaitsgory and Preygel), there is a Morita equivalence between Q(X x_Y X) and Q(Y), that is, an equivalence of their infinity-categories of stable module infinity-categories. In particular, this immediately implies an identification of their Drinfeld centers (as previously established in 0805.0157), and more generally, an identification of their associated topological field theories. Another consequence is that for an affine algebraic group G in characteristic zero and an algebraic subgroup K, passing to K-invariants induces an equivalence from stable infinity-categories with algebraic G-action to modules for the Hecke category Q(K\G/K).

Factorization homology theories of topological manifolds, after Beilinson, Drinfeld and Lurie, are homology-type theories for topological $n$-manifolds whose coefficient systems are $n$-disk algebras or $n$-disk stacks. In this work we prove a precise formulation of this idea, giving an axiomatic characterization of factorization homology with coefficients in $n$-disk algebras in terms of a generalization of the Eilenberg--Steenrod axioms for singular homology. Each such theory gives rise to a kind of topological quantum field theory, for which observables can be defined on general $n$-manifolds and not only closed $n$-manifolds. For $n$-disk algebra coefficients, these field theories are characterized by the condition that global observables are determined by local observables in a strong sense. Our axiomatic point of view has a number of applications. In particular, we give a concise proof of the nonabelian Poincar\'e duality of Salvatore, Segal, and Lurie. We present some essential classes of calculations of factorization homology, such as for free $n$-disk algebras and enveloping algebras of Lie algebras, several of which have a conceptual meaning in terms of Koszul duality., Comment: 37 pages. Corrected Proposition 5.3

We provide a framework for the study of structured manifolds with singularities and their locally determined invariants. This generalizes factorization homology, or topological chiral homology, to the setting of singular manifolds equipped with various tangential structures. Examples of such factorization homology theories include intersection homology, compactly supported stratified mapping spaces, and Hochschild homology with coefficients. Factorization homology theories for singular manifolds are characterized by a generalization of the Eilenberg-Steenrod axioms. Using these axioms, we extend the nonabelian Poincar\'e duality of Salvatore and Lurie to the setting of singular manifolds -- this is a nonabelian version of the Poincar\'e duality given by intersection homology. We pay special attention to the simple case of singular manifolds whose singularity datum is a properly embedded submanifold and give a further simplified algebraic characterization of these homology theories. In the case of 3-manifolds with 1-dimensional submanifolds, this structure gives rise to knot and link homology theories akin to Khovanov homology., Comment: This pre-print is now superseded by the two pre-prints arXiv:1409.0848 and 1409.0501

In this work, we study the deformation theory of $\cE_n$-rings and the $\cE_n$ analogue of the tangent complex, or topological Andr\'e-Quillen cohomology. We prove a generalization of a conjecture of Kontsevich, that there is a fiber sequence $A[n-1] \ra T_A\ra \hh^*_{\cE_{n}}(A)[n]$, relating the $\cE_n$-tangent complex and $\cE_n$-Hochschild cohomology of an $\cE_n$-ring $A$. We give two proofs: The first is direct, reducing the problem to certain stable splittings of configuration spaces of punctured Euclidean spaces; the second is more conceptual, where we identify the sequence as the Lie algebras of a fiber sequence of derived algebraic groups, $B^{n-1}A^\times\ra \Aut_A\ra \Aut_{\fB^nA}$. Here $\fB^nA$ is an enriched $(\oo,n)$-category constructed from $A$, and $\cE_n$-Hochschild cohomology is realized as the infinitesimal automorphisms of $\fB^nA$. These groups are associated to moduli problems in $\cE_{n+1}$-geometry, a {\it less} commutative form of derived algebraic geometry, in the sense of To\"en-Vezzosi and Lurie. Applying techniques of Koszul duality, this sequence consequently attains a nonunital $\cE_{n+1}$-algebra structure; in particular, the shifted tangent complex $T_A[-n]$ is a nonunital $\cE_{n+1}$-algebra. The $\cE_{n+1}$-algebra structure of this sequence extends the previously known $\cE_{n+1}$-algebra structure on $\hh^*_{\cE_{n}}(A)$, given in the higher Deligne conjecture. In order to establish this moduli-theoretic interpretation, we make extensive use of factorization homology, a homology theory for framed $n$-manifolds with coefficients given by $\cE_n$-algebras, constructed as a topological analogue of Beilinson-Drinfeld's chiral homology. We give a separate exposition of this theory, developing the necessary results used in our proofs. This work is an elaboration of a chapter of the author's 2008 PhD thesis, \cite{thez}., Comment: May vary slightly from the published version

We extend the theory of chiral and factorization algebras, developed for curves by Beilinson and Drinfeld in \cite{bd}, to higher-dimensional varieties. This extension entails the development of the homotopy theory of chiral and factorization structures, in a sense analogous to Quillen's homotopy theory of differential graded Lie algebras. We prove the equivalence of higher-dimensional chiral and factorization algebras by embedding factorization algebras into a larger category of chiral commutative coalgebras, then realizing this interrelation as a chiral form of Koszul duality. We apply these techniques to rederive some fundamental results of \cite{bd} on chiral enveloping algebras of $\star$-Lie algebras.

We study the interaction between geometric operations on stacks and algebraic operations on their categories of sheaves. We work in the general setting of derived algebraic geometry: our basic objects are derived stacks X and their oo-categories QC(X) of quasicoherent sheaves. We show that for a broad class of derived stacks, called perfect stacks, algebraic and geometric operations on their categories of sheaves are compatible. We identify the category of sheaves on a fiber product with the tensor product of the categories of sheaves on the factors. We also identify the category of sheaves on a fiber product with functors between the categories of sheaves on the factors (thus realizing functors as integral transforms, generalizing a theorem of Toen for ordinary schemes). As a first application, for a perfect stack X, consider QC(X) with its usual monoidal tensor product. Then our main results imply the equivalence of the Drinfeld center (or Hochschild cohomology category) of QC(X), the trace (or Hochschild homology category) of QC(X) and the category of sheaves on the loop space of X. More generally, we show that the E_n-center and the E_n-trace (or E_n-Hochschild cohomology and homology categories respectively) of QC(X) are equivalent to the category of sheaves on the space of maps from the n-sphere into X. This directly verifies geometric instances of the categorified Deligne and Kontsevich conjectures on the structure of Hochschild cohomology. As a second application, we use our main results to calculate the Drinfeld center of categories of linear endofunctors of categories of sheaves. This provides concrete applications to the structure of Hecke algebras in geometric representation theory. Finally, we explain how all of the above results can be interpreted in the context of topological field theory., Comment: Final revised version, to appear, Jour. Amer. Math. Soc

The Robotic Refueling Mission 3 (RRM3) payload was installed on the International Space Station (ISS) in December 2018. Its Cryogenic Demonstation Subsystem (CDS) was loaded with 19 kilograms of liquid methane, for demonstration of on-orbit storage and transfer. We provide a short description of the design and ground testing of the CDS, and then discuss the on-orbit operations through the first half of 2019.

Liquid hydrogen (LH2) and liquid oxygen (LO2) cryogenic propellants can dramatically enhance NASA sability to explore the solar system due to their superior specific impulse (Isp) capability. Although these cryogenic propellants can be challenging to manage and store, they allow significant mass advantages over traditional hypergolic propulsion systems and are therefore enabling for many planetary science missions. New cryogenic storage techniques such as sub-cooling and the use of advanced insulation and low thermal conductivity support structures will allow for the long term storage and use of cryogenic propellants for solar system exploration and hence allow NASA to deliver more payloads to targets of interest, launch on smaller and less expensive launch vehicles, or both. These new cryogenic storage technologies were implemented in a design study for the Titan Orbiter Polar Surveyor (TOPS) mission, with LH2 and LO2 as propellants, and the resulting spacecraft design was able to achieve a 43 launch mass reduction over a TOPS mission, that utilized a traditional hypergolic propulsion system with monomethylhydrazine (MMH) and nitrogen tetroxide (NTO) propellants. This paper describes the cryogenic propellant storage design for the TOPS mission and demonstrates how these cryogenic propellants are stored passively for a decade-long Titan mission that requires the cryogenics propellants to be stored for 8.5 years.

Cryogenic propellants such as liquid hydrogen (LH2) and liquid oxygen (LOX) can dramatically enhance NASAs ability to explore the solar system because of their superior specific impulse (Isp) capability. Although these cryogenic propellants can be challenging to manage and store, they allow significant mass advantages over traditional hypergolic propulsion systems and are therefore technically enabling for many planetary science missions. New cryogenic storage techniques such as subcooling and the use of advanced insulation and low thermal conductivity support structures will allow for the long term storage and use of cryogenic propellants for solar system exploration and hence allow NASA to deliver more payloads to targets of interest, launch on smaller and less expensive launch vehicles, or both. Employing cryogenic propellants will allow NASA to perform missions to planetary destinations that would not be possible with the use of traditional hypergolic propellants. These new cryogenic storage technologies were implemented in a design study for the Titan Orbiter Polar Surveyor (TOPS) mission, with LH2 and LOX as propellants, and the resulting spacecraft design was able to achieve a 43 launch mass reduction over a TOPS mission, that utilized a conventional hypergolic propulsion system with mono-methyl hydrazine (MMH) and nitrogen tetroxide (NTO) propellants. This paper describes the cryogenic propellant storage design for the TOPS mission and demonstrates how these cryogenic propellants are stored passively for a decade-long Titan mission.

Cryogenic propellants such as liquid hydrogen (LH2) and liquid oxygen (LOX) can dramatically enhance NASAs ability to explore the solar system because of their superior specific impulse (Isp) capability. Although these cryogenic propellants can be challenging to manage and store, they allow significant mass advantages over traditional hypergolic propulsion systems and are therefore technically enabling for many planetary science missions. New cryogenic storage techniques such as subcooling and the use of advanced insulation and low thermal conductivity support structures will allow for the long term storage and use of cryogenic propellants for solar system exploration and hence allow NASA to deliver more payloads to targets of interest, launch on smaller and less expensive launch vehicles, or both. Employing cryogenic propellants will allow NASA to perform missions to planetary destinations that would not be possible with the use of traditional hypergolic propellants. These new cryogenic storage technologies were implemented in a design study for the Titan Orbiter Polar Surveyor (TOPS) mission, with LH2 and LOX as propellants, and the resulting spacecraft design was able to achieve a 43 launch mass reduction over a TOPS mission, that utilized a conventional hypergolic propulsion system with mono-methyl hydrazine (MMH) and nitrogen tetroxide (NTO) propellants. This paper describes the cryogenic propellant storage design for the TOPS mission and demonstrates how these cryogenic propellants are stored passively for a decade-long Titan mission.

Titanium 15-3-3-3 and Kevlar 49 are highly desired materials for structural components in cryogenic applications due to their low thennal conductivity at low temperatures. Previous tests have indicated that titanium 15-3-3-3 becomes increasingly brittle as the temperature decreases. Furthermore, little is known regarding the mechanical properties of Kevlar 49 at low temperatures, most specifically its Young's modulus. This testing investigates the mechanical properties of both materials at cryogenic temperatures through cryogenic mechanical tensile testing to failure. The elongation, ultimate tensile strength, yield strength, and break strength of both materials are provided and analyzed here.

This shroud provides a deep-space simulating environment for testing scaled-down models of passively cooling systems for spaceflight optics and instruments. It is used inside a liquid-nitrogen- cooled vacuum chamber, and it is cooled by liquid helium to 5 K. It has an inside geometry of approximately 1.6 m diameter by 0.45 m tall. The inside surfaces of its top and sidewalls have a thermal absorptivity greater than 0.96. The bottom wall has a large central opening that is easily customized to allow a specific test item to extend through it. This enables testing of scale models of realistic passive cooling configurations that feature a very large temperature drop between the deepspace-facing cooled side and the Sun/Earth-facing warm side. This shroud has an innovative thermal closeout of the bottom wall, so that a test sample can have a hot (room temperature) side outside of the shroud, and a cold side inside the shroud. The combination of this closeout and the very black walls keeps radiated heat from the sample s warm end from entering the shroud, reflecting off the walls and heating the sample s cold end. The shroud includes 12 vertical rectangular sheet-copper side panels that are oriented in a circular pattern. Using tabs bent off from their edges, these side panels are bolted to each other and to a steel support ring on which they rest. The removable shroud top is a large copper sheet that rests on, and is bolted to, the support ring when the shroud is closed. The support ring stands on four fiberglass tube legs, which isolate it thermally from the vacuum chamber bottom. The insides of the cooper top and side panels are completely covered with 25- mm-thick aluminum honeycomb panels. This honeycomb is painted black before it is epoxied to the copper surfaces. A spiral-shaped copper tube, clamped at many different locations to the outside of the top copper plate, serves as part of the liquid helium cooling loop. Another copper tube, plumbed in a series to the top plate s tube, is clamped to the sidewall tabs where they are bolted to the support ring. Flowing liquid helium through these tubes cools the entire shroud to 5 K. The entire shroud is wrapped loosely in a layer of double-aluminized Kapton. The support ring s inner diameter is the largest possible hole through which the test item can extend into the shroud. Twelve custom-sized trapezoidal copper sheets extend inward from the support ring to within a few millimeters of the test item. Attached to the inner edge of each of these sheets is a custom-shaped strip of Kapton, which is aluminum- coated on the warm-facing (outer) side, and has thin Dacron netting attached to its cold-facing side. This Kapton rests against the test item, but the Dacron keeps it from making significant thermal contact. The result is a non-contact, radiatively reflective thermal closeout with essentially no gap through which radiation can pass. In this way, the part of the test item outside the shroud can be heated to relatively high temperatures without any radiative heat leaking to the inside.

A compact and lightweight mechanism has been developed to accurately move a Fourier transform spectrometer (FTS) scan mirror (a cube corner) in a near-linear fashion with near constant speed at cryogenic temperatures. This innovation includes a slide mechanism to restrict motion to one dimension, an actuator to drive the motion, and a linear velocity transducer (LVT) to measure the speed. The cube corner mirror is double-passed in one arm of the FTS; double-passing is required to compensate for optical beam shear resulting from tilting of the moving cube corner. The slide, actuator, and LVT are off-the-shelf components that are capable of cryogenic vacuum operation. The actuator drives the slide for the required travel of 2.5 cm. The LVT measures translation speed. A proportional feedback loop compares the LVT voltage with the set voltage (speed) to derive an error signal to drive the actuator and achieve near constant speed. When the end of the scan is reached, a personal computer reverses the set voltage. The actuator and LVT have no moving parts in contact, and have magnetic properties consistent with cryogenic operation. The unlubricated slide restricts motion to linear travel, using crossed roller bearings consistent with 100-million- stroke operation. The mechanism tilts several arc seconds during transport of the FTS mirror, which would compromise optical fringe efficiency when using a flat mirror. Consequently, a cube corner mirror is used, which converts a tilt into a shear. The sheared beam strikes (at normal incidence) a flat mirror at the end of the FTS arm with the moving mechanism, thereby returning upon itself and compensating for the shear

A compact cryogenic test chamber can be cooled to approximately 5 to 6 Kelvin for materials testing. The system includes a temperature controller and multiple sensors to measure specimen temperature at different locations. The testing chamber provides a fast and easy method to perform materials testing at lower than liquid nitrogen temperature (77 K). The purpose of the chamber is to cool a composite lap shear specimen to approximately 20 K so that tensile test force and displacement data may be acquired at this cryogenic temperature range.

Constellation-X is a multi-satellite x-ray astronomy mission presently being planned for launch in the 2010 time frame. Each of 4 identical satellites will contain a telescope and instruments for imaging and spectroscopic analysis of both hard and soft x-rays. The X- ray Microcalorimeter Spectrometer (XMS) instrument will use arrays of microcalorimeters to detect X-rays with energies from 0.2 to 6 keV. The technologies under development for this instrument include Transition-Edge Sensors (TES) with multiplexed SQUID readouts and NTD-Ge detectors with JFET readouts. Both will be operated at temperatures in the 50-60 mK range and both have a projected cooling power requirement of approximately 5 microwatts. In addition, in order to meet the lifetime requirement of 6 years (with a goal of 10 years), a mechanical cryocooler will be used to provide a heat sink for the low temperature cooler. The required performance is 20 mW at 6 K, with a goal of 4 K operation. In this paper we present the development status of an adiabatic demagnetization refigerator (ADR) that meets the cooling requirements of the X M S instrument. At present we have demonstrated a 4-stage ADR that operates continuously at 50 mK using a 4.2 K helium bath as a heat sink. The cooling power is 21 microwatts at 100 mK and 6 microwatts at 50 mK. Its efficiency when operating at 50 mK is 1 1 % of Carnot (accounting for all dissipation at 4.2 K and below, but not including power dissipation in the room temperature electronics), but this is expected to rise to 25% in the next generation system in which active gas-gap heat switches are replaced with passive devices. This will reduce the peak heat rejection rate of the ADR to less than 7 mW at 6 K. Details of the ADR s design and operation, as well as the development program leading up to a flight-qualified instrument, will be discussed.

The effectiveness of vocational education at New York State's public two-year colleges was investigated during spring 1976. Part one of the project was a follow-up study of 1970-71 and 1973-74 graduates to ascertain their success in job or transfer placement within fields for which they were trained, salary and job mobility patterns, and opinions on their college educational preparation and job satisfaction. Part two acquired information on employers' satisfaction with graduates' training, work performance, and relationship with campus placement offices. The study indicated graduates were generally satisfied with their college major choices, seemingly influenced by the image of a career in that field. Graduates felt that courses in their major were rigid and difficult although also stimulating and challenging. Future employment success and advancement seemed hinged on a need for further education; those continuing their education did so mostly on a part-time basis, in programs related to their college majors. Graduate location data indicated the existence of a local job market at least for the first job after graduation. College placement offices were felt to be more helpful in supplying job listings and forwarding credentials than in job counseling. Employers were generally satisfied with graduates' college preparation but believed the liaison between colleges and employers could be improved. (Author/TR)

The Faculty Grants for the Improvement of Undergraduate Instruction, Excellence in Teaching Awards, and Distinguished Teaching Professorships programs were implemented in 1972 by the State University of New York (SUNY) in an effort to recognize and reward teaching excellence. An evaluation of these programs was conducted as part of a national assessment project sponsored by the Department of Health, Education and Welfare. The study included surveys of 1,500 faculty members to determine their awareness of and reaction to the programs, and a survey of program participants concerning the selection process and program impact. In addition, interviews were conducted at nine SUNY campuses with program administrators, committee members and participants concerning program implementation. Sample items are appended. Although each group expressed general approval of the programs, 85% of the faculty felt that they had little or no impact upon the importance accorded teaching. Program participants also felt that the most marked result of participation was heightened personal satisfaction, and administrators and committee members felt that factors irrelevant to teaching excellence had a bearing upon award decisions. The dissemination of information among departments, administration and campuses is a major barrier to the effectiveness of these programs. Suggestions for program improvements are offered. (Author/JAC)