Showing 1–50 of 50 results for author: Powell, G

Relative nonhomogeneous Koszul duality for PROPs associated to nonaugmented operads

Authors: Geoffrey Powell

Abstract: The purpose of this paper is to show how Positselski's relative nonhomogeneous Koszul duality theory applies when studying the linear category underlying the PROP associated to a (non-augmented) operad of a certain form, in particular assuming that the reduced part of the operad is binary quadratic. In this case, the linear category has both a left augmentation and a right augmentation (correspond… ▽ More The purpose of this paper is to show how Positselski's relative nonhomogeneous Koszul duality theory applies when studying the linear category underlying the PROP associated to a (non-augmented) operad of a certain form, in particular assuming that the reduced part of the operad is binary quadratic. In this case, the linear category has both a left augmentation and a right augmentation (corresponding to different units), using Positselski's terminology. The general theory provides two associated linear differential graded (DG) categories; indeed, in this framework, one can work entirely within the DG realm, as opposed to the curved setting required for Positselski's general theory. Moreover, DG modules over DG categories are related by adjunctions. When the reduced part of the operad is Koszul (working over a field of characteristic zero), the relative Koszul duality theory shows that there is a Koszul-type equivalence between the appropriate homotopy categories of DG modules. This gives a form of Koszul duality relationship between the above DG categories. This is illustrated by the case of the operad encoding unital, commutative associative algebras, extending the classical Koszul duality between commutative associative algebras and Lie algebras. In this case, the associated linear category is the linearization of the category of finite sets and all maps. The relative nonhomogeneous Koszul duality theory relates its derived category to the respective homotopy categories of modules over two explicit linear DG categories. △ Less

Submitted 12 June, 2024; originally announced June 2024.

Comments: 52 pages. Comments welcome

CattleFace-RGBT: RGB-T Cattle Facial Landmark Benchmark

Authors: Ethan Coffman, Reagan Clark, Nhat-Tan Bui, Trong Thang Pham, Beth Kegley, Jeremy G. Powell, Jiangchao Zhao, Ngan Le

Abstract: To address this challenge, we introduce CattleFace-RGBT, a RGB-T Cattle Facial Landmark dataset consisting of 2,300 RGB-T image pairs, a total of 4,600 images. Creating a landmark dataset is time-consuming, but AI-assisted annotation can help. However, applying AI to thermal images is challenging due to suboptimal results from direct thermal training and infeasible RGB-thermal alignment due to dif… ▽ More To address this challenge, we introduce CattleFace-RGBT, a RGB-T Cattle Facial Landmark dataset consisting of 2,300 RGB-T image pairs, a total of 4,600 images. Creating a landmark dataset is time-consuming, but AI-assisted annotation can help. However, applying AI to thermal images is challenging due to suboptimal results from direct thermal training and infeasible RGB-thermal alignment due to different camera views. Therefore, we opt to transfer models trained on RGB to thermal images and refine them using our AI-assisted annotation tool following a semi-automatic annotation approach. Accurately localizing facial key points on both RGB and thermal images enables us to not only discern the cattle's respiratory signs but also measure temperatures to assess the animal's thermal state. To the best of our knowledge, this is the first dataset for the cattle facial landmark on RGB-T images. We conduct benchmarking of the CattleFace-RGBT dataset across various backbone architectures, with the objective of establishing baselines for future research, analysis, and comparison. The dataset and models are at https://github.com/UARK-AICV/CattleFace-RGBT-benchmark △ Less

Submitted 5 June, 2024; originally announced June 2024.

Stochastic optimization with arbitrary recurrent data sampling

Authors: William G. Powell, Hanbaek Lyu

Abstract: For obtaining optimal first-order convergence guarantee for stochastic optimization, it is necessary to use a recurrent data sampling algorithm that samples every data point with sufficient frequency. Most commonly used data sampling algorithms (e.g., i.i.d., MCMC, random reshuffling) are indeed recurrent under mild assumptions. In this work, we show that for a particular class of stochastic optim… ▽ More For obtaining optimal first-order convergence guarantee for stochastic optimization, it is necessary to use a recurrent data sampling algorithm that samples every data point with sufficient frequency. Most commonly used data sampling algorithms (e.g., i.i.d., MCMC, random reshuffling) are indeed recurrent under mild assumptions. In this work, we show that for a particular class of stochastic optimization algorithms, we do not need any other property (e.g., independence, exponential mixing, and reshuffling) than recurrence in data sampling algorithms to guarantee the optimal rate of first-order convergence. Namely, using regularized versions of Minimization by Incremental Surrogate Optimization (MISO), we show that for non-convex and possibly non-smooth objective functions, the expected optimality gap converges at an optimal rate $O(n^{-1/2})$ under general recurrent sampling schemes. Furthermore, the implied constant depends explicitly on the `speed of recurrence', measured by the expected amount of time to visit a given data point either averaged (`target time') or supremized (`hitting time') over the current location. We demonstrate theoretically and empirically that convergence can be accelerated by selecting sampling algorithms that cover the data set most effectively. We discuss applications of our general framework to decentralized optimization and distributed non-negative matrix factorization. △ Less

Submitted 15 January, 2024; originally announced January 2024.

Comments: 41 pages, 3 figures, 1 table

On fundamental structure underlying Lie algebra homology with coefficients tensor products of the adjoint representation

Authors: Geoffrey Powell

Abstract: This paper exhibits fundamental structure underlying Lie algebra homology with coefficients in tensor products of the adjoint representation, mostly focusing upon the case of free Lie algebras. The main result yields a DG category that is constructed from the PROP associated to the Lie operad. Underlying this is a two-term complex of bimodules over this PROP; it is a quotient of the universal Ch… ▽ More This paper exhibits fundamental structure underlying Lie algebra homology with coefficients in tensor products of the adjoint representation, mostly focusing upon the case of free Lie algebras. The main result yields a DG category that is constructed from the PROP associated to the Lie operad. Underlying this is a two-term complex of bimodules over this PROP; it is a quotient of the universal Chevalley-Eilenberg complex. The homology of this DG category is intimately related to outer functors over free groups (introduced in earlier joint work with Vespa). This uses the author's previous results relating functors on free groups to representations of the PROP associated to the Lie operad. This gives a direct algebraic explanation as to why the degree one homology should correspond to an outer functor. Hitherto, the only known argument relied upon the relationship with the higher Hochschild homology functors that arise from the work of Turchin and Willwacher. △ Less

Submitted 14 September, 2023; originally announced September 2023.

Comments: 18 pages. Comments welcome

On the Passi and the Mal'cev functors

Authors: Geoffrey Powell

Abstract: The author has shown that the category of analytic contravariant functors on $\mathbf{gr}$, the category of finitely-generated free groups, is equivalent to the category of left modules over the PROP associated to the Lie operad, working over $\mathbb{Q}$. This exploited properties of the polynomial filtration of the category of contravariant functors on $\mathbf{gr}$. The first purpose of this… ▽ More The author has shown that the category of analytic contravariant functors on $\mathbf{gr}$, the category of finitely-generated free groups, is equivalent to the category of left modules over the PROP associated to the Lie operad, working over $\mathbb{Q}$. This exploited properties of the polynomial filtration of the category of contravariant functors on $\mathbf{gr}$. The first purpose of this paper is to strengthen the corresponding result for covariant functors on $\mathbf{gr}$. This involves introducing the appropriate analogue of the category of analytic contravariant functors, namely a certain category of towers of polynomial functors on $\mathbf{gr}$. This category is abelian and has a natural symmetric monoidal structure induced by the usual tensor product of functors. Moreover, the projective generators of this category are described in terms of the Mal'cev functors that are introduced here. It follows that this category is equivalent to the category of right modules over the PROP associated to the Lie operad. As a fundamental example, the Passi functors arising from the group ring functors are described explicitly. The theory is applied to consider bifunctors on $\mathbf{gr}$. This allows the $\mathbb{Q}$-linearization of the category of free groups to be described, up to polynomial filtration. As a stronger application of the theory, this is generalized to the Casimir PROP associated to the Lie operad, as studied by Hinich and Vaintrob. Up to polynomial filtration, this recovers the category $\mathbf{A}$ introduced by Habiro and Massuyeau in their study of bottom tangles in handlebodies. △ Less

Submitted 14 September, 2023; originally announced September 2023.

Comments: 43 pages. Comments welcome

"It's Weird That it Knows What I Want": Usability and Interactions with Copilot for Novice Programmers

Authors: James Prather, Brent N. Reeves, Paul Denny, Brett A. Becker, Juho Leinonen, Andrew Luxton-Reilly, Garrett Powell, James Finnie-Ansley, Eddie Antonio Santos

Abstract: Recent developments in deep learning have resulted in code-generation models that produce source code from natural language and code-based prompts with high accuracy. This is likely to have profound effects in the classroom, where novices learning to code can now use free tools to automatically suggest solutions to programming exercises and assignments. However, little is currently known about how… ▽ More Recent developments in deep learning have resulted in code-generation models that produce source code from natural language and code-based prompts with high accuracy. This is likely to have profound effects in the classroom, where novices learning to code can now use free tools to automatically suggest solutions to programming exercises and assignments. However, little is currently known about how novices interact with these tools in practice. We present the first study that observes students at the introductory level using one such code auto-generating tool, Github Copilot, on a typical introductory programming (CS1) assignment. Through observations and interviews we explore student perceptions of the benefits and pitfalls of this technology for learning, present new observed interaction patterns, and discuss cognitive and metacognitive difficulties faced by students. We consider design implications of these findings, specifically in terms of how tools like Copilot can better support and scaffold the novice programming experience. △ Less

Submitted 5 April, 2023; originally announced April 2023.

Comments: 26 pages, 2 figures, TOCHI

Improving Multimodal Interactive Agents with Reinforcement Learning from Human Feedback

Authors: Josh Abramson, Arun Ahuja, Federico Carnevale, Petko Georgiev, Alex Goldin, Alden Hung, Jessica Landon, Jirka Lhotka, Timothy Lillicrap, Alistair Muldal, George Powell, Adam Santoro, Guy Scully, Sanjana Srivastava, Tamara von Glehn, Greg Wayne, Nathaniel Wong, Chen Yan, Rui Zhu

Abstract: An important goal in artificial intelligence is to create agents that can both interact naturally with humans and learn from their feedback. Here we demonstrate how to use reinforcement learning from human feedback (RLHF) to improve upon simulated, embodied agents trained to a base level of competency with imitation learning. First, we collected data of humans interacting with agents in a simulate… ▽ More An important goal in artificial intelligence is to create agents that can both interact naturally with humans and learn from their feedback. Here we demonstrate how to use reinforcement learning from human feedback (RLHF) to improve upon simulated, embodied agents trained to a base level of competency with imitation learning. First, we collected data of humans interacting with agents in a simulated 3D world. We then asked annotators to record moments where they believed that agents either progressed toward or regressed from their human-instructed goal. Using this annotation data we leveraged a novel method - which we call "Inter-temporal Bradley-Terry" (IBT) modelling - to build a reward model that captures human judgments. Agents trained to optimise rewards delivered from IBT reward models improved with respect to all of our metrics, including subsequent human judgment during live interactions with agents. Altogether our results demonstrate how one can successfully leverage human judgments to improve agent behaviour, allowing us to use reinforcement learning in complex, embodied domains without programmatic reward functions. Videos of agent behaviour may be found at https://youtu.be/v_Z9F2_eKk4. △ Less

Submitted 21 November, 2022; originally announced November 2022.

Baby bead representations

Authors: Geoffrey Powell

Abstract: This paper is motivated by the study of Turchin and Willwacher's bead representations. The problem is reformulated here in terms of the Lie algebra homology of a free Lie algebra with coefficients in tensor products of the adjoint representation. The main idea is to exploit the truncation of the coefficients given by killing Lie brackets of length greater than two. Although this truncation is br… ▽ More This paper is motivated by the study of Turchin and Willwacher's bead representations. The problem is reformulated here in terms of the Lie algebra homology of a free Lie algebra with coefficients in tensor products of the adjoint representation. The main idea is to exploit the truncation of the coefficients given by killing Lie brackets of length greater than two. Although this truncation is brutal, it retains significant and highly non-trivial information, as exhibited by explicit results. A dévissage is used that splits the problem into two steps, separating out a `homology' calculation from `antisymmetrization'. This involves some auxiliary categories, including a generalization of the upper walled Brauer category. This approach passes through the `baby bead representations' of the title, for which complete results are obtained. As an application, the composition factors of Turchin and Willwacher's bead representations are calculated for a new infinite family. △ Less

Submitted 19 September, 2022; originally announced September 2022.

Comments: 59 pages

On the FI-homology of the injective cogenerators

Authors: Geoffrey Powell

Abstract: The purpose of this paper is to give information on the FI-homology of the standard injective cogenerators of the category of FI-modules, where FI is the category of finite sets and injections. Working over a field k of characteristic zero, a full calculation is given in homological degree zero and a conjectural description in higher homological degree. The proof of the main theorem reduces to… ▽ More The purpose of this paper is to give information on the FI-homology of the standard injective cogenerators of the category of FI-modules, where FI is the category of finite sets and injections. Working over a field k of characteristic zero, a full calculation is given in homological degree zero and a conjectural description in higher homological degree. The proof of the main theorem reduces to a calculation in representation theory of the symmetric groups, exploiting the Young orthonormal basis. △ Less

Submitted 19 September, 2022; originally announced September 2022.

Comments: 22 pages

Numerical reconstruction for 3D nonlinear SAR imaging via a version of the convexification method

Authors: Vo Anh Khoa, Michael Victor Klibanov, William Grayson Powell, Loc Hoang Nguyen

Abstract: This work extends the applicability of our recent convexification-based algorithm for constructing images of the dielectric constant of buried or occluded target. We are orientated towards the detection of explosive-like targets such as antipersonnel land mines and improvised explosive devices in the non-invasive inspections of buildings. In our previous work, the method is posed in the perspectiv… ▽ More This work extends the applicability of our recent convexification-based algorithm for constructing images of the dielectric constant of buried or occluded target. We are orientated towards the detection of explosive-like targets such as antipersonnel land mines and improvised explosive devices in the non-invasive inspections of buildings. In our previous work, the method is posed in the perspective that we use multiple source locations running along a line of source to get a 2D image of the dielectric function. Mathematically, we solve a 1D coefficient inverse problem for a hyperbolic equation for each source location. Different from any conventional Born approximation-based technique for synthetic-aperture radar, this method does not need any linearization. In this paper, we attempt to verify the method using several 3D numerical tests with simulated data. We revisit the global convergence of the gradient descent method of our computational approach. △ Less

Submitted 19 June, 2022; originally announced June 2022.

Comments: 23 pages, 11 figures

MSC Class: 78A46; 65L70; 65C20

Outer functors and a general operadic framework

Authors: Geoffrey Powell

Abstract: For $\mathcal{O}$ an operad in $k$-vector spaces, the category $\mathcal{F}_\mathcal{O}$ is defined to be the category of $k$-linear functors from the PROP associated to $\mathcal{O}$ to $k$-vector spaces. Given $μ\in \mathcal{O} (2)$ that satisfies a right Leibniz condition, the full subcategory $\mathcal{F}_\mathcal{O}^μ\subset \mathcal{F}_\mathcal{O}$ is introduced here and its properties studi… ▽ More For $\mathcal{O}$ an operad in $k$-vector spaces, the category $\mathcal{F}_\mathcal{O}$ is defined to be the category of $k$-linear functors from the PROP associated to $\mathcal{O}$ to $k$-vector spaces. Given $μ\in \mathcal{O} (2)$ that satisfies a right Leibniz condition, the full subcategory $\mathcal{F}_\mathcal{O}^μ\subset \mathcal{F}_\mathcal{O}$ is introduced here and its properties studied. This is motivated by the case of the Lie operad, where $μ$ is taken to be the generator. By previous results of the author, when $k = \mathbb{Q}$, $\mathcal{F}_{Lie}$ is equivalent to the category of analytic functors on the opposite of the category $\mathbf{gr}$ of finitely-generated free groups. The main result shows that $\mathcal{F}_{Lie}^μ$ identifies with the category of outer analytic functors, as introduced in earlier work of the author with Vespa. Using this identification, this theory has applications to the study of the higher Hochschild homology functors related to work of Turchin and Willwacher. △ Less

Submitted 26 October, 2023; v1 submitted 31 January, 2022; originally announced January 2022.

Comments: v3: minor revision; now 20 pages. v2: updated presentation, with some improvements. Main results unchanged. 18 pages

On analytic contravariant functors on free groups

Authors: Geoffrey Powell

Abstract: Working over a field $k$ of characteristic zero, the category of analytic contravariant functors on the category of finitely-generated free groups is shown to be equivalent to the category of representations of the $k$-linear category associated to the Lie operad. Two proofs are given of this result. The first uses the original Ginzburg-Kapranov approach to Koszul duality of binary quadratic ope… ▽ More Working over a field $k$ of characteristic zero, the category of analytic contravariant functors on the category of finitely-generated free groups is shown to be equivalent to the category of representations of the $k$-linear category associated to the Lie operad. Two proofs are given of this result. The first uses the original Ginzburg-Kapranov approach to Koszul duality of binary quadratic operads and the fact that the category of analytic contravariant functors is Koszul. The second proof proceeds by making the equivalence explicit using the $k$-linear category associated to the operad encoding unital associative algebras, which provides the `twisting bimodule'. A key ingredient is the Poincaré-Birkhoff-Witt theorem. Using the explicit formulation, it is shown how this equivalence reflects the tensor product on the category of analytic contravariant functors, relating this to the convolution product for representations of the category associated to the Lie operad. △ Less

Submitted 25 May, 2023; v1 submitted 5 October, 2021; originally announced October 2021.

Comments: (v2) 34 pages: some improvements to the exposition. (v1) 29 pages

Homological splitting results for modules over Leibniz algebras

Authors: Geoffrey Powell

Abstract: A unified splitting result for Ext calculated in the category of modules over a Leibniz algebra is given for the case where coefficients are either both symmetric modules or both antisymmetric modules. This is a generalization of results of Loday and Pirashvili and others. A unified splitting result for Ext calculated in the category of modules over a Leibniz algebra is given for the case where coefficients are either both symmetric modules or both antisymmetric modules. This is a generalization of results of Loday and Pirashvili and others. △ Less

Submitted 22 September, 2021; originally announced September 2021.

Comments: 18 pages

The primitive filtration of the Leibniz complex

Authors: Geoffrey Powell

Abstract: Pirashvili exhibited a small subcomplex of the Leibniz complex $(T(s \mathfrak{g}), d_{\mathrm{Leib}})$ of a Leibniz algebra $\mathfrak{g}$. The main result of this paper generalizes this result to show that the primitive filtration of $T(s\mathfrak{g})$ provides an increasing, exhaustive filtration of the Leibniz complex by subcomplexes, thus establishing a conjecture due to Loday. The associated… ▽ More Pirashvili exhibited a small subcomplex of the Leibniz complex $(T(s \mathfrak{g}), d_{\mathrm{Leib}})$ of a Leibniz algebra $\mathfrak{g}$. The main result of this paper generalizes this result to show that the primitive filtration of $T(s\mathfrak{g})$ provides an increasing, exhaustive filtration of the Leibniz complex by subcomplexes, thus establishing a conjecture due to Loday. The associated spectral sequence is used to give a new proof of Pirashvili's conjecture that, when $\mathfrak{g}$ is a free Leibniz algebra, the homology of the Pirashvili complex is zero except in degree one. This result is then used to show that the desuspension of the Pirashvili complex carries a natural $L_\infty$-structure that induces the natural Lie algebra structure on the homology of the complex in degree zero. △ Less

Submitted 25 October, 2022; v1 submitted 22 September, 2021; originally announced September 2021.

Comments: 21 pages: revised version, following suggestions of a referee (accepted). (v1: 18 pages.)

On derivations of free algebras over operads and the generalized divergence

Authors: Geoffrey Powell

Abstract: For $\mathcal{O}$ a reduced operad, a generalized divergence from the derivations of a free $\mathcal{O}$-algebra to a suitable trace space is constructed. In the case of the Lie operad, this corresponds to Satoh's trace map and, for the associative operad, to the double divergence of Alekseev, Kawazumi, Kuno and Naef. The generalized divergence is shown to be a $1$-cocycle for the usual Lie algeb… ▽ More For $\mathcal{O}$ a reduced operad, a generalized divergence from the derivations of a free $\mathcal{O}$-algebra to a suitable trace space is constructed. In the case of the Lie operad, this corresponds to Satoh's trace map and, for the associative operad, to the double divergence of Alekseev, Kawazumi, Kuno and Naef. The generalized divergence is shown to be a $1$-cocycle for the usual Lie algebra structure on derivations. These results place the previous constructions into a unified framework; moreover, they are natural with respect to the operad. An important new ingredient is the use of naturality with respect to the category of finite-rank free modules and split monomorphisms over a commutative ring $R$. This allows the notion of torsion for such functors to be exploited. Supposing that the ring $R$ is a PID and that the operad $\mathcal{O}$ is binary, the main result relates the kernel of the generalized divergence to the sub Lie algebra of the Lie algebra of derivations that is generated by the elements of degree one with respect to the grading induced by arity. △ Less

Submitted 19 May, 2021; originally announced May 2021.

Comments: 41 pages

Supervised multi-specialist topic model with applications on large-scale electronic health record data

Authors: Ziyang Song, Xavier Sumba Toral, Yixin Xu, Aihua Liu, Liming Guo, Guido Powell, Aman Verma, David Buckeridge, Ariane Marelli, Yue Li

Abstract: Motivation: Electronic health record (EHR) data provides a new venue to elucidate disease comorbidities and latent phenotypes for precision medicine. To fully exploit its potential, a realistic data generative process of the EHR data needs to be modelled. We present MixEHR-S to jointly infer specialist-disease topics from the EHR data. As the key contribution, we model the specialist assignments a… ▽ More Motivation: Electronic health record (EHR) data provides a new venue to elucidate disease comorbidities and latent phenotypes for precision medicine. To fully exploit its potential, a realistic data generative process of the EHR data needs to be modelled. We present MixEHR-S to jointly infer specialist-disease topics from the EHR data. As the key contribution, we model the specialist assignments and ICD-coded diagnoses as the latent topics based on patient's underlying disease topic mixture in a novel unified supervised hierarchical Bayesian topic model. For efficient inference, we developed a closed-form collapsed variational inference algorithm to learn the model distributions of MixEHR-S. We applied MixEHR-S to two independent large-scale EHR databases in Quebec with three targeted applications: (1) Congenital Heart Disease (CHD) diagnostic prediction among 154,775 patients; (2) Chronic obstructive pulmonary disease (COPD) diagnostic prediction among 73,791 patients; (3) future insulin treatment prediction among 78,712 patients diagnosed with diabetes as a mean to assess the disease exacerbation. In all three applications, MixEHR-S conferred clinically meaningful latent topics among the most predictive latent topics and achieved superior target prediction accuracy compared to the existing methods, providing opportunities for prioritizing high-risk patients for healthcare services. MixEHR-S source code and scripts of the experiments are freely available at https://github.com/li-lab-mcgill/mixehrS △ Less

Submitted 3 May, 2021; originally announced May 2021.

A Pirashvili-type theorem for functors on non-empty finite sets

Authors: Geoffrey Powell, Christine Vespa

Abstract: Pirashvili's Dold-Kan type theorem for finite pointed sets follows from the identification in terms of surjections of the morphisms between the tensor powers of a functor playing the role of the augmentation ideal; these functors are projective. We give an unpointed analogue of this result: namely, we compute the morphisms between the tensor powers of the corresponding functor in the unpointed con… ▽ More Pirashvili's Dold-Kan type theorem for finite pointed sets follows from the identification in terms of surjections of the morphisms between the tensor powers of a functor playing the role of the augmentation ideal; these functors are projective. We give an unpointed analogue of this result: namely, we compute the morphisms between the tensor powers of the corresponding functor in the unpointed context. We also calculate the Ext groups between such objects, in particular showing that these functors are not projective; this is an important difference between the pointed and unpointed contexts. This work is motivated by our functorial analysis of the higher Hochschild homology of a wedge of circles. △ Less

Submitted 23 September, 2020; originally announced September 2020.

Solving Rubik's Cube with a Robot Hand

Authors: OpenAI, Ilge Akkaya, Marcin Andrychowicz, Maciek Chociej, Mateusz Litwin, Bob McGrew, Arthur Petron, Alex Paino, Matthias Plappert, Glenn Powell, Raphael Ribas, Jonas Schneider, Nikolas Tezak, Jerry Tworek, Peter Welinder, Lilian Weng, Qiming Yuan, Wojciech Zaremba, Lei Zhang

Abstract: We demonstrate that models trained only in simulation can be used to solve a manipulation problem of unprecedented complexity on a real robot. This is made possible by two key components: a novel algorithm, which we call automatic domain randomization (ADR) and a robot platform built for machine learning. ADR automatically generates a distribution over randomized environments of ever-increasing di… ▽ More We demonstrate that models trained only in simulation can be used to solve a manipulation problem of unprecedented complexity on a real robot. This is made possible by two key components: a novel algorithm, which we call automatic domain randomization (ADR) and a robot platform built for machine learning. ADR automatically generates a distribution over randomized environments of ever-increasing difficulty. Control policies and vision state estimators trained with ADR exhibit vastly improved sim2real transfer. For control policies, memory-augmented models trained on an ADR-generated distribution of environments show clear signs of emergent meta-learning at test time. The combination of ADR with our custom robot platform allows us to solve a Rubik's cube with a humanoid robot hand, which involves both control and state estimation problems. Videos summarizing our results are available: https://openai.com/blog/solving-rubiks-cube/ △ Less

Submitted 15 October, 2019; originally announced October 2019.

Emergent Tool Use From Multi-Agent Autocurricula

Authors: Bowen Baker, Ingmar Kanitscheider, Todor Markov, Yi Wu, Glenn Powell, Bob McGrew, Igor Mordatch

Abstract: Through multi-agent competition, the simple objective of hide-and-seek, and standard reinforcement learning algorithms at scale, we find that agents create a self-supervised autocurriculum inducing multiple distinct rounds of emergent strategy, many of which require sophisticated tool use and coordination. We find clear evidence of six emergent phases in agent strategy in our environment, each of… ▽ More Through multi-agent competition, the simple objective of hide-and-seek, and standard reinforcement learning algorithms at scale, we find that agents create a self-supervised autocurriculum inducing multiple distinct rounds of emergent strategy, many of which require sophisticated tool use and coordination. We find clear evidence of six emergent phases in agent strategy in our environment, each of which creates a new pressure for the opposing team to adapt; for instance, agents learn to build multi-object shelters using moveable boxes which in turn leads to agents discovering that they can overcome obstacles using ramps. We further provide evidence that multi-agent competition may scale better with increasing environment complexity and leads to behavior that centers around far more human-relevant skills than other self-supervised reinforcement learning methods such as intrinsic motivation. Finally, we propose transfer and fine-tuning as a way to quantitatively evaluate targeted capabilities, and we compare hide-and-seek agents to both intrinsic motivation and random initialization baselines in a suite of domain-specific intelligence tests. △ Less

Submitted 10 February, 2020; v1 submitted 16 September, 2019; originally announced September 2019.

Atomic Modeling of Photoionization Fronts in Nitrogen Gas

Authors: William J. Gray, P. A. Keiter, H. Lefevre, C. R. Patterson, J. S. Davis, K. G. Powell, C. C. Kuranz, R. P. Drake

Abstract: Photoionization fronts play a dominant role in many astrophysical environments, but remain difficult to achieve in a laboratory experiment. Recent papers have suggested that experiments using a nitrogen medium held at ten atmospheres of pressure that is irradiated by a source with a radiation temperature of T$_{\rm R}\sim$ 100 eV can produce viable photoionization fronts. We present a suite of one… ▽ More Photoionization fronts play a dominant role in many astrophysical environments, but remain difficult to achieve in a laboratory experiment. Recent papers have suggested that experiments using a nitrogen medium held at ten atmospheres of pressure that is irradiated by a source with a radiation temperature of T$_{\rm R}\sim$ 100 eV can produce viable photoionization fronts. We present a suite of one-dimensional numerical simulations using the \helios\ multi-material radiation hydrodynamics code that models these conditions and the formation of a photoionization front. We study the effects of varying the atomic kinetics and radiative transfer model on the hydrodynamics and ionization state of the nitrogen gas, finding that more sophisticated physics, in particular a multi-angle long characteristic radiative transfer model and a collisional-radiative atomics model, dramatically changes the atomic kinetic evolution of the gas. A photoionization front is identified by computing the ratios between the photoionization rate, the electron impact ionization rate, and the total recombination rate. We find that due to the increased electron temperatures found using more advanced physics that photoionization fronts are likely to form in our nominal model. We report results of several parameter studies. In one of these, the nitrogen pressure is fixed at ten atmospheres and varies the source radiation temperature while another fixes the temperature at 100 eV and varied the nitrogen pressure. Lower nitrogen pressures increase the likelihood of generating a photoionization front while varying the peak source temperature has little effect. △ Less

Submitted 18 April, 2019; originally announced April 2019.

Comments: 17 pages, 10 figures, accepted to physics of plasmas

Modeling disease progression in longitudinal EHR data using continuous-time hidden Markov models

Authors: Aman Verma, Guido Powell, Yu Luo, David Stephens, David L. Buckeridge

Abstract: Modeling disease progression in healthcare administrative databases is complicated by the fact that patients are observed only at irregular intervals when they seek healthcare services. In a longitudinal cohort of 76,888 patients with chronic obstructive pulmonary disease (COPD), we used a continuous-time hidden Markov model with a generalized linear model to model healthcare utilization events. W… ▽ More Modeling disease progression in healthcare administrative databases is complicated by the fact that patients are observed only at irregular intervals when they seek healthcare services. In a longitudinal cohort of 76,888 patients with chronic obstructive pulmonary disease (COPD), we used a continuous-time hidden Markov model with a generalized linear model to model healthcare utilization events. We found that the fitted model provides interpretable results suitable for summarization and hypothesis generation. △ Less

Submitted 2 December, 2018; originally announced December 2018.

Comments: Machine Learning for Health (ML4H) Workshop at NeurIPS 2018 arXiv:1811.07216

Report number: ML4H/2018/145

Symmetric powers, Steenrod operations and representation stability

Authors: Geoffrey Powell

Abstract: Working over the prime field F_p, the structure of the indecomposables Q^* for the action of the algebra of Steenrod reduced powers A(p) on the symmetric power functors S^* is studied by exploiting the theory of strict polynomial functors. In particular, working at the prime 2, representation stability is exhibited for certain related functors, leading to a conjectural representation stability d… ▽ More Working over the prime field F_p, the structure of the indecomposables Q^* for the action of the algebra of Steenrod reduced powers A(p) on the symmetric power functors S^* is studied by exploiting the theory of strict polynomial functors. In particular, working at the prime 2, representation stability is exhibited for certain related functors, leading to a conjectural representation stability description of quotients of Q^* arising from the polynomial filtration of symmetric powers. △ Less

Submitted 13 February, 2019; v1 submitted 24 September, 2018; originally announced September 2018.

Comments: Revised version (with modified title), focusing upon the core material; now 20 pages

MSC Class: 55S10; 20G43

Learning Dexterous In-Hand Manipulation

Authors: OpenAI, Marcin Andrychowicz, Bowen Baker, Maciek Chociej, Rafal Jozefowicz, Bob McGrew, Jakub Pachocki, Arthur Petron, Matthias Plappert, Glenn Powell, Alex Ray, Jonas Schneider, Szymon Sidor, Josh Tobin, Peter Welinder, Lilian Weng, Wojciech Zaremba

Abstract: We use reinforcement learning (RL) to learn dexterous in-hand manipulation policies which can perform vision-based object reorientation on a physical Shadow Dexterous Hand. The training is performed in a simulated environment in which we randomize many of the physical properties of the system like friction coefficients and an object's appearance. Our policies transfer to the physical robot despite… ▽ More We use reinforcement learning (RL) to learn dexterous in-hand manipulation policies which can perform vision-based object reorientation on a physical Shadow Dexterous Hand. The training is performed in a simulated environment in which we randomize many of the physical properties of the system like friction coefficients and an object's appearance. Our policies transfer to the physical robot despite being trained entirely in simulation. Our method does not rely on any human demonstrations, but many behaviors found in human manipulation emerge naturally, including finger gaiting, multi-finger coordination, and the controlled use of gravity. Our results were obtained using the same distributed RL system that was used to train OpenAI Five. We also include a video of our results: https://youtu.be/jwSbzNHGflM △ Less

Submitted 18 January, 2019; v1 submitted 1 August, 2018; originally announced August 2018.

Comments: Making OpenAI the first author. We wish this paper to be cited as "Learning Dexterous In-Hand Manipulation" by OpenAI et al. We are replicating the approach from the physics community: arXiv:1812.06489

Laboratory Photoionization Fronts in Nitrogen Gas: A Numerical Feasibility and Parameter Study

Authors: William J Gray, P. A. Keiter, H. Lefevre, C. R. Patterson, J. S. Davis, B. van Der Holst, K. G. Powell, R. P. Drake

Abstract: Photoionization fronts play a dominant role in many astrophysical situations, but remain difficult to achieve in a laboratory experiment. We present the results from a computational parameter study evaluating the feasibility of the photoionization experiment presented in the design paper by Drake, R. P., Hazak, G., Keiter, P. A., Davis, J. S., Patterson, C. R., Frank, A., Blackman, E. G., & Busque… ▽ More Photoionization fronts play a dominant role in many astrophysical situations, but remain difficult to achieve in a laboratory experiment. We present the results from a computational parameter study evaluating the feasibility of the photoionization experiment presented in the design paper by Drake, R. P., Hazak, G., Keiter, P. A., Davis, J. S., Patterson, C. R., Frank, A., Blackman, E. G., & Busquet, M. 2016, ApJ, 833, 249 in which a photoionization front is generated in a nitrogen medium . The nitrogen gas density and the Planckian radiation temperature of the x-ray source define each simulation. Simulations modeled experiments in which the x-ray flux is generated by a laser-heated gold foil, suitable for experiments using many kJ of laser energy, and experiments in which the flux is generated by a "z-pinch" device, which implodes a cylindrical shell of conducting wires. The models are run using CRASH, our block-adaptive-mesh code for multi-material radiation hydrodynamics. The radiative transfer model uses multi-group, flux-limited diffusion with thirty radiation groups. In addition, electron heat conduction is modeled using a single-group, flux-limited diffusion. In the theory, a photoionization front can exist only when the ratios of the electron recombination rate to the photoionization rate and the electron impact ionization rate to the recombination rate lie in certain ranges. These ratios are computed for several ionization states of nitrogen. Photoionization fronts are found to exist for laser driven models with moderate nitrogen densities ($\sim$10$^{21}$ cm$^{-3}$) and radiation temperatures above 90 eV. For "z-pinch" driven models, lower nitrogen densities are preferred ($<$10$^{21}$ cm$^{-3}$). We conclude that the proposed experiments are likely to generate photoionization fronts. △ Less

Submitted 4 April, 2018; originally announced April 2018.

Comments: 22 pages, 14 figures, comments welcome

Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research

Authors: Matthias Plappert, Marcin Andrychowicz, Alex Ray, Bob McGrew, Bowen Baker, Glenn Powell, Jonas Schneider, Josh Tobin, Maciek Chociej, Peter Welinder, Vikash Kumar, Wojciech Zaremba

Abstract: The purpose of this technical report is two-fold. First of all, it introduces a suite of challenging continuous control tasks (integrated with OpenAI Gym) based on currently existing robotics hardware. The tasks include pushing, sliding and pick & place with a Fetch robotic arm as well as in-hand object manipulation with a Shadow Dexterous Hand. All tasks have sparse binary rewards and follow a Mu… ▽ More The purpose of this technical report is two-fold. First of all, it introduces a suite of challenging continuous control tasks (integrated with OpenAI Gym) based on currently existing robotics hardware. The tasks include pushing, sliding and pick & place with a Fetch robotic arm as well as in-hand object manipulation with a Shadow Dexterous Hand. All tasks have sparse binary rewards and follow a Multi-Goal Reinforcement Learning (RL) framework in which an agent is told what to do using an additional input. The second part of the paper presents a set of concrete research ideas for improving RL algorithms, most of which are related to Multi-Goal RL and Hindsight Experience Replay. △ Less

Submitted 10 March, 2018; v1 submitted 26 February, 2018; originally announced February 2018.

Higher Hochschild homology and exponential functors

Authors: Geoffrey Powell, Christine Vespa

Abstract: We study higher Hochschild homology evaluated on wedges of circles, viewed as a functor on the category of free groups. The principal results use coefficients arising from square-zero extensions; this is motivated in part by work of Turchin and Willwacher in relation to hairy graph cohomology. Working over a field of characteristic zero, achieve effective calculations of higher Hochschild homolo… ▽ More We study higher Hochschild homology evaluated on wedges of circles, viewed as a functor on the category of free groups. The principal results use coefficients arising from square-zero extensions; this is motivated in part by work of Turchin and Willwacher in relation to hairy graph cohomology. Working over a field of characteristic zero, achieve effective calculations of higher Hochschild homology, generalizing initial results of Turchin and Willwacher. This exploits exponential functors, which arise both in calculating higher Hochschild homology and in relation to polynomial functors on free groups. It is a crucial observation that these are related. For the main cases of interest, the action of automorphisms of free groups on Hochschild homology factors across outer automorphisms. As the appropriate underlying framework, we introduce and make essential use of the category of outer functors, the full subcategory of functors on free groups on which inner automorphisms act trivially. △ Less

Submitted 15 October, 2019; v1 submitted 21 February, 2018; originally announced February 2018.

Comments: v3 Further minor revision; 96 pages. (v2: Minor revision; now 96 pages.)

MSC Class: 55N99; 18A25; 13D03; 20J05

Finite presheaves and $A$-finite generation of unstable algebras mod nilpotents

Authors: Geoffrey Powell

Abstract: Inspired by the work of Henn, Lannes and Schwartz on unstable algebras over the Steenrod algebra modulo nilpotents, a characterization of unstable algebras that are $A$-finitely generated up to nilpotents is given in terms of the associated presheaf, by introducing the notion of a finite presheaf. In particular, this gives the natural characterization of the (co)analytic presheaves that are import… ▽ More Inspired by the work of Henn, Lannes and Schwartz on unstable algebras over the Steenrod algebra modulo nilpotents, a characterization of unstable algebras that are $A$-finitely generated up to nilpotents is given in terms of the associated presheaf, by introducing the notion of a finite presheaf. In particular, this gives the natural characterization of the (co)analytic presheaves that are important in the theory of Henn, Lannes and Schwartz. However, finite presheaves remain imperfectly understood, as illustrated by examples. One important class of examples is shown to be provided by unstable algebras of finite transcendence degree (under a necessary weak finiteness condition). For unstable Hopf algebras, it is shown that the situation is much better: the associated presheaf is finite if and only if its growth function is polynomial. This leads to a description of unstable Hopf algebras modulo nilpotents in the spirit of Henn, Lannes and Schwartz. △ Less

Submitted 21 June, 2018; v1 submitted 24 July, 2017; originally announced July 2017.

Comments: Slightly streamlined version, now 24 pages

Algebraic infinite deloo** and derived destabilization

Authors: Geoffrey Powell

Abstract: Working over the prime field of characteristic two, consequences of the Koszul duality between the Steenrod algebra and the big Dyer-Lashof algebra are studied, with an emphasis on the interplay between instability for the Steenrod algebra action and that for the Dyer-Lashof operations. The central algebraic framework is the category of length-graded modules over the Steenrod algebra equipped with… ▽ More Working over the prime field of characteristic two, consequences of the Koszul duality between the Steenrod algebra and the big Dyer-Lashof algebra are studied, with an emphasis on the interplay between instability for the Steenrod algebra action and that for the Dyer-Lashof operations. The central algebraic framework is the category of length-graded modules over the Steenrod algebra equipped with an unstable action of the Dyer-Lashof algebra, with compatibility via the Nishida relations. A first ingredient is a functor defined on modules over the Steenrod algebra that arose in the work of Kuhn and McCarty on the homology of infinite loop spaces. This functor is given in terms of derived functors of destabilization from the category of modules over the Steenrod algebra to unstable modules, enriched by taking into account the action of Dyer-Lashof operations. A second ingredient is the derived functors of the Dyer-Lashof indecomposables functor to length-graded modules over the Steenrod algebra. These are related to functors used by Miller in his study of a spectral sequence to calculate the homology of an infinite deloo**. An important fact is that these functors can be calculated as the homology of an explicit Koszul complex with terms expressed as certain Steinberg functors. The latter are quadratic dual to the more familiar Singer functors. By exploiting the explicit complex built from the Singer functors which calculates the derived functors of destabilization, Koszul duality leads to an algebraic infinite deloo** spectral sequence. This is conceptually similar to Miller's spectral sequence, but there seems to be no direct relationship. The spectral sequence sheds light on the relationship between unstable modules over the Steenrod algebra and all modules. △ Less

Submitted 11 May, 2017; v1 submitted 23 January, 2017; originally announced January 2017.

Comments: Minor revision with no changes to mathematical content; 36 pages (v1 38 pages)

MSC Class: 55S10

The A-decomposability of the Singer construction

Authors: Nguyen H. V. Hung, Geoffrey Powell

Abstract: Let $R_s M$ denote the Singer construction on an unstable module $M$ over the Steenrod algebra $A$ at the prime two; $R_s M$ is canonically a subobject of $P_s\otimes M$, where $P_s$ is the polynomial algebra on s generators of degree one. Passage to $A$-indecomposables gives the natural transformation $R_s M \rightarrow F \otimes_A (P_s \otimes M)$, which identifies with the dual of the compositi… ▽ More Let $R_s M$ denote the Singer construction on an unstable module $M$ over the Steenrod algebra $A$ at the prime two; $R_s M$ is canonically a subobject of $P_s\otimes M$, where $P_s$ is the polynomial algebra on s generators of degree one. Passage to $A$-indecomposables gives the natural transformation $R_s M \rightarrow F \otimes_A (P_s \otimes M)$, which identifies with the dual of the composition of the Singer transfer and the Lannes-Zarati homomorphism. The main result of the paper proves the weak generalized algebraic spherical class conjecture, which was proposed by the first named author. Namely, this morphism is trivial on elements of positive degree when s>2. The condition s>2 is necessary, as exhibited by the spherical classes of Hopf invariant one and those of Kervaire invariant one. △ Less

Submitted 27 September, 2018; v1 submitted 30 June, 2016; originally announced June 2016.

Comments: v2 15 pages. Minor revision. v3 17 pages, revision following referee's recommendations. Accepted for publication J. Alg

MSC Class: 55S10; 18E10

On double Poisson structures on commutative algebras

Authors: Geoffrey Powell

Abstract: Double Poisson structures (a la Van den Bergh) on commutative algebras are studied; the main result shows that there are no non-trivial such structures on polynomial algebras of Krull dimension greater than one. For a general commutative algebra A, this places significant restrictions on possible double Poisson structures. Exotic double Poisson structures are exhibited by the case of the polynomia… ▽ More Double Poisson structures (a la Van den Bergh) on commutative algebras are studied; the main result shows that there are no non-trivial such structures on polynomial algebras of Krull dimension greater than one. For a general commutative algebra A, this places significant restrictions on possible double Poisson structures. Exotic double Poisson structures are exhibited by the case of the polynomial algebra on a single generator, previously considered by Van den Bergh. △ Less

Submitted 8 July, 2016; v1 submitted 24 March, 2016; originally announced March 2016.

Comments: 12 pages; very minor revision

MSC Class: 17B63

Essential extensions, the nilpotent filtration and the Arone-Goodwillie tower

Authors: Geoffrey Powell

Abstract: The spectral sequence associated to the Arone-Goodwillie tower for the n-fold loop space functor is used to show that the first two non-trivial layers of the nilpotent filtration of the reduced mod 2 cohomology of a (sufficiently connected) space with nilpotent cohomology are comparable. This relies upon the theory of unstable modules over the mod 2 Steenrod algebra, together with properties of a… ▽ More The spectral sequence associated to the Arone-Goodwillie tower for the n-fold loop space functor is used to show that the first two non-trivial layers of the nilpotent filtration of the reduced mod 2 cohomology of a (sufficiently connected) space with nilpotent cohomology are comparable. This relies upon the theory of unstable modules over the mod 2 Steenrod algebra, together with properties of a generalized class of almost unstable modules which is introduced here. An essential ingredient of the proof is a non-vanishing result for certain extension groups in the category of unstable modules localized away from nilpotents. △ Less

Submitted 24 February, 2016; v1 submitted 10 May, 2015; originally announced May 2015.

Comments: Minor revision. Sketch of case p odd added. This version 27 pages

On the derived functors of destabilization and of iterated loop functors

Authors: Geoffrey Powell

Abstract: These notes explain how to construct small functorial chain complexes which calculate the derived functors of destabilization (respectively iterated loop functors) in the theory of modules over the mod 2 Steenrod algebra; this shows how to unify results of Singer and of Lannes and Zarati. These notes explain how to construct small functorial chain complexes which calculate the derived functors of destabilization (respectively iterated loop functors) in the theory of modules over the mod 2 Steenrod algebra; this shows how to unify results of Singer and of Lannes and Zarati. △ Less

Submitted 17 July, 2017; v1 submitted 30 March, 2015; originally announced March 2015.

Comments: Minor revision; more detail provided on the Lannes-Zarati morphism and a new section giving perspectives. 32 pages. (v2 Minor revision. 25 pages.)

Truncated projective spaces, Brown-Gitler spectra and indecomposable A(1)-modules

Authors: Geoffrey Powell

Abstract: A structure theorem for bounded-below modules over the subalgebra A(1) of the mod 2 Steenrod algebra generated by Sq^1, Sq^2 is proved; this is applied to prove a classification theorem for a family of indecomposable A(1)-modules. The action of the A(1)-Picard group on this family is described, as is the behaviour of duality. The cohomology of dual Brown-Gitler spectra is identified within this… ▽ More A structure theorem for bounded-below modules over the subalgebra A(1) of the mod 2 Steenrod algebra generated by Sq^1, Sq^2 is proved; this is applied to prove a classification theorem for a family of indecomposable A(1)-modules. The action of the A(1)-Picard group on this family is described, as is the behaviour of duality. The cohomology of dual Brown-Gitler spectra is identified within this family and the relation with members of the A(1)-Picard group is made explicit. Similarly, the cohomology of truncated projective spaces is considered within this classification; this leads to a conceptual understanding of various results within the literature. In particular, a unified approach to Ext-groups relevant to Adams spectral sequence calculations is obtained, englobing earlier results of Davis (for truncated projective spaces) and recent work of Pearson (for Brown-Gitler spectra). △ Less

Submitted 26 December, 2014; v1 submitted 21 April, 2014; originally announced April 2014.

Comments: Revision - mathematical content essentially unchanged (39 pages)

MSC Class: 19L41; 55S10

Around conjectures of N. Kuhn

Authors: Nguyen The Cuong, Gérald Gaudens, Geoffrey Powell, Lionel Schwartz

Abstract: We discuss two extensions of results conjectured by Nick Kuhn about the non-realization of unstable algebras as the mod $p$ singular cohomology of a space, for $p$ a prime. The first extends and refines earlier work of the second and fourth authors, using Lannes' map** space theorem. The second (for the prime $2$) is based on an analysis of the $-1$ and $-2$ columns of the Eilenberg-Moore spect… ▽ More We discuss two extensions of results conjectured by Nick Kuhn about the non-realization of unstable algebras as the mod $p$ singular cohomology of a space, for $p$ a prime. The first extends and refines earlier work of the second and fourth authors, using Lannes' map** space theorem. The second (for the prime $2$) is based on an analysis of the $-1$ and $-2$ columns of the Eilenberg-Moore spectral sequence, and of the associated extension. In both cases, the statements and proofs use the relationship between the categories of unstable modules and functors between $\Fp$-vector spaces. The second result in particular exhibits the power of the functorial approach. △ Less

Submitted 5 February, 2015; v1 submitted 11 February, 2014; originally announced February 2014.

On the BP<n>-cohomology of elementary abelian p-groups

Authors: Geoffrey Powell

Abstract: The structure of the BP<n>-cohomology of elementary abelian p-groups is studied, obtaining a presentation expressed in terms of BP-cohomology and mod-p singular cohomology, using the Milnor derivations. The arguments are based on a result on multi-Koszul complexes which is related to Margolis's criterion for freeness of a graded module over an exterior algebra. The structure of the BP<n>-cohomology of elementary abelian p-groups is studied, obtaining a presentation expressed in terms of BP-cohomology and mod-p singular cohomology, using the Milnor derivations. The arguments are based on a result on multi-Koszul complexes which is related to Margolis's criterion for freeness of a graded module over an exterior algebra. △ Less

Submitted 13 January, 2015; v1 submitted 3 April, 2013; originally announced April 2013.

Comments: Minor revision (18 pages)

MSC Class: 55N20; 55N22; 20J06

On connective KO-theory of elementary abelian 2-groups

Authors: Geoffrey Powell

Abstract: A general notion of detection is introduced and used in the study of the cohomology of elementary abelian 2-groups with respect to the spectra in the Postnikov tower of orthogonal K-theory. This recovers and extends results of Bruner and Greenlees and is related to calculations of the (co)homology of the spaces of the associated Omega-spectra by Stong and by Cowen Morton. A general notion of detection is introduced and used in the study of the cohomology of elementary abelian 2-groups with respect to the spectra in the Postnikov tower of orthogonal K-theory. This recovers and extends results of Bruner and Greenlees and is related to calculations of the (co)homology of the spaces of the associated Omega-spectra by Stong and by Cowen Morton. △ Less

Submitted 10 February, 2014; v1 submitted 30 July, 2012; originally announced July 2012.

MSC Class: 19L41; 20J06

Journal ref: Algebr. Geom. Topol. 14 (2014) 2693-2720

Simulating the long-term evolution of radiative shocks in shock tubes

Authors: B. van der Holst, G. Toth, I. V. Sokolov, B. R. Torralva, K. G. Powell, R. P. Drake

Abstract: We present the latest improvements in the Center for Radiative Shock Hydrodynamics (CRASH) code, a parallel block-adaptive-mesh Eulerian code for simulating high-energy-density plasmas. The implementation can solve for radiation models with either a gray or a multigroup method in the flux-limited-diffusion approximation. The electrons and ions are allowed to be out of temperature equilibrium and f… ▽ More We present the latest improvements in the Center for Radiative Shock Hydrodynamics (CRASH) code, a parallel block-adaptive-mesh Eulerian code for simulating high-energy-density plasmas. The implementation can solve for radiation models with either a gray or a multigroup method in the flux-limited-diffusion approximation. The electrons and ions are allowed to be out of temperature equilibrium and flux-limited electron thermal heat conduction is included. We have recently implemented a CRASH laser package with 3-D ray tracing, resulting in improved energy deposition evaluation. New, more accurate opacity models are available which significantly improve radiation transport in materials like xenon. In addition, the HYPRE preconditioner has been added to improve the radiation implicit solver. With this updated version of the CRASH code we study radiative shock tube problems. In our set-up, a 1 ns, 3.8 kJ laser pulse irradiates a 20 micron beryllium disk, driving a shock into a xenon-filled plastic tube. The electrons emit radiation behind the shock. This radiation from the shocked xenon preheats the unshocked xenon. Photons traveling ahead of the shock will also interact with the plastic tube, heat it, and in turn this can drive another shock off the wall into the xenon. We are now able to simulate the long term evolution of radiative shocks. △ Less

Submitted 6 June, 2012; originally announced June 2012.

Comments: submitted to High Energy Density Physics

On connective K-theory of elementary abelian 2-groups and local duality

Authors: Geoffrey Powell

Abstract: The connective ku-(co)homology of elementary abelian 2-groups is determined as a functor of the elementary abelian 2-group. The argument requires only the calculation of the rank one case and the Atiyah-Segal theorem for KU-cohomology together with an analysis of the functorial structure of the integral group ring. The methods can also be applied to the odd primary case. These results are used t… ▽ More The connective ku-(co)homology of elementary abelian 2-groups is determined as a functor of the elementary abelian 2-group. The argument requires only the calculation of the rank one case and the Atiyah-Segal theorem for KU-cohomology together with an analysis of the functorial structure of the integral group ring. The methods can also be applied to the odd primary case. These results are used to analyse the local cohomology spectral sequence calculating ku-homology, via a functorial version of local duality for Koszul complexes. This gives a conceptual explanation of results of Bruner and Greenlees. △ Less

Submitted 29 December, 2011; originally announced December 2011.

MSC Class: 19L41; 20J06

Simulating radiative shocks in nozzle shock tubes

Authors: B. van der Holst, G. Toth, I. V. Sokolov, L. K. S. Daldorff, K. G. Powell, R. P. Drake

Abstract: We use the recently developed Center for Radiative Shock Hydrodynamics (CRASH) code to numerically simulate laser-driven radiative shock experiments. These shocks are launched by an ablated beryllium disk and are driven down xenon-filled plastic tubes. The simulations are initialized by the two-dimensional version of the Lagrangian Hyades code which is used to evaluate the laser energy deposition… ▽ More We use the recently developed Center for Radiative Shock Hydrodynamics (CRASH) code to numerically simulate laser-driven radiative shock experiments. These shocks are launched by an ablated beryllium disk and are driven down xenon-filled plastic tubes. The simulations are initialized by the two-dimensional version of the Lagrangian Hyades code which is used to evaluate the laser energy deposition during the first 1.1ns. The later times are calculated with the CRASH code. This code solves for the multi-material hydrodynamics with separate electron and ion temperatures on an Eulerian block-adaptive-mesh and includes a multi-group flux-limited radiation diffusion and electron thermal heat conduction. The goal of the present paper is to demonstrate the capability to simulate radiative shocks of essentially three-dimensional experimental configurations, such as circular and elliptical nozzles. We show that the compound shock structure of the primary and wall shock is captured and verify that the shock properties are consistent with order-of-magnitude estimates. The produced synthetic radiographs can be used for comparison with future nozzle experiments at high-energy-density laser facilities. △ Less

Submitted 18 September, 2011; originally announced September 2011.

Comments: submitted to High Energy Density Physics

Crash: A Block-Adaptive-Mesh Code for Radiative Shock Hydrodynamics - Implementation and Verification

Authors: B. van der Holst, G. Toth, I. V. Sokolov, K. G. Powell, J. P. Holloway, E. S. Myra, Q. Stout, M. L. Adams, J. E. Morel, R. P. Drake

Abstract: We describe the CRASH (Center for Radiative Shock Hydrodynamics) code, a block adaptive mesh code for multi-material radiation hydrodynamics. The implementation solves the radiation diffusion model with the gray or multigroup method and uses a flux limited diffusion approximation to recover the free-streaming limit. The electrons and ions are allowed to have different temperatures and we include a… ▽ More We describe the CRASH (Center for Radiative Shock Hydrodynamics) code, a block adaptive mesh code for multi-material radiation hydrodynamics. The implementation solves the radiation diffusion model with the gray or multigroup method and uses a flux limited diffusion approximation to recover the free-streaming limit. The electrons and ions are allowed to have different temperatures and we include a flux limited electron heat conduction. The radiation hydrodynamic equations are solved in the Eulerian frame by means of a conservative finite volume discretization in either one, two, or three-dimensional slab geometry or in two-dimensional cylindrical symmetry. An operator split method is used to solve these equations in three substeps: (1) solve the hydrodynamic equations with shock-capturing schemes, (2) a linear advection of the radiation in frequency-logarithm space, and (3) an implicit solve of the stiff radiation diffusion, heat conduction, and energy exchange. We present a suite of verification test problems to demonstrate the accuracy and performance of the algorithms. The CRASH code is an extension of the Block-Adaptive Tree Solarwind Roe Upwind Scheme (BATS-R-US) code with this new radiation transfer and heat conduction library and equation-of-state and multigroup opacity solvers. Both CRASH and BATS-R-US are part of the publicly available Space Weather Modeling Framework (SWMF). △ Less

Submitted 19 January, 2011; originally announced January 2011.

Comments: 51 pages, 19 figures; submitted to Astrophysical Journal

On quadratic coalgebras, duality and the universal Steenrod algebra

Authors: Geoffrey Powell

Abstract: The notion of quadratic self-duality for coalgebras is developed with applications to algebraic structures which arise naturally in algebraic topology, related to the universal Steenrod algebra via an appropriate form of duality. This explains and unifies results of Lomonaco and Singer. The notion of quadratic self-duality for coalgebras is developed with applications to algebraic structures which arise naturally in algebraic topology, related to the universal Steenrod algebra via an appropriate form of duality. This explains and unifies results of Lomonaco and Singer. △ Less

Submitted 31 December, 2010; originally announced January 2011.

Comments: 18 pages

MSC Class: 55S10

On the derived functors of destabilization at odd primes

Authors: Geoffrey Powell

Abstract: An explicit chain complex is constructed to calculate the derived functors of destabilization at an odd prime, generalizing constructions of Zarati and of Hung and Sum. The methods are based on the ideas of Singer and Miller and also apply at the prime two. A structural result on the derived functors of destabilization is deduced. An explicit chain complex is constructed to calculate the derived functors of destabilization at an odd prime, generalizing constructions of Zarati and of Hung and Sum. The methods are based on the ideas of Singer and Miller and also apply at the prime two. A structural result on the derived functors of destabilization is deduced. △ Less

Submitted 28 February, 2013; v1 submitted 31 December, 2010; originally announced January 2011.

Comments: Significant revision (27 pages); accepted for publication in Acta Mathematica Vietnamica (AMV)

MSC Class: 55S10

On the double transfer and the f-invariant

Authors: Geoffrey Powell

Abstract: The purpose of this paper is to investigate an algebraic version of the double complex transfer, in particular the classes in the two-line of the Adams-Novikov spectral sequence which are the image of comodule primitives of the MU-homology of the product of two copies of infinite complex projective space via the algebraic double transfer. These classes are analysed by two related approaches; the f… ▽ More The purpose of this paper is to investigate an algebraic version of the double complex transfer, in particular the classes in the two-line of the Adams-Novikov spectral sequence which are the image of comodule primitives of the MU-homology of the product of two copies of infinite complex projective space via the algebraic double transfer. These classes are analysed by two related approaches; the first, p-locally for an odd prime, by using the morphism induced in MU-homology by the chromatic factorization of the double transfer map together with the f'-invariant of Behrens (for p>=5). The second approach uses the algebraic double transfer and the f-invariant of Laures. △ Less

Submitted 9 March, 2011; v1 submitted 6 May, 2010; originally announced May 2010.

Comments: 25 pages (revised)

MSC Class: 55R12

On the Singer functor R_1 and the functor Fix

Authors: Geoffrey Powell

Abstract: Lannes' T-functor is used to give a construction of the Singer functor R_1 on the category U of unstable modules over the Steenrod algebra A. This leads to a direct proof that the composite functor Fix R_1 is naturally equivalent to the identity. Further properties of the functors R_1 are deduced, especially when applied to reduced and nilclosed unstable modules. Lannes' T-functor is used to give a construction of the Singer functor R_1 on the category U of unstable modules over the Steenrod algebra A. This leads to a direct proof that the composite functor Fix R_1 is naturally equivalent to the identity. Further properties of the functors R_1 are deduced, especially when applied to reduced and nilclosed unstable modules. △ Less

Submitted 4 December, 2009; originally announced December 2009.

MSC Class: 55S10

Embedding the flag representation in divided powers

Authors: Geoffrey Powell

Abstract: A generalization of a theorem of Crabb and Hubbuck concerning the embedding of flag representations in divided powers is given, working over an arbitrary finite field F, using the category of functors from finite-dimensional F-vector spaces to F-vector spaces. A generalization of a theorem of Crabb and Hubbuck concerning the embedding of flag representations in divided powers is given, working over an arbitrary finite field F, using the category of functors from finite-dimensional F-vector spaces to F-vector spaces. △ Less

Submitted 18 September, 2009; originally announced September 2009.

Comments: Accepted for publication in JHRS

MSC Class: 55S10

Subrings of singular cohomology associated to spectra

Authors: Geoffrey M L Powell

Abstract: This paper extends the relation established for group cohomology by Green, Hunton and Schuster between chromatic phenomena in stable homotopy theory and certain natural subrings of singular cohomology. This exploits the theory due to Henn, Lannes and Schwartz of unstable algebras over the Steenrod algebra localized away from nilpotents. This paper extends the relation established for group cohomology by Green, Hunton and Schuster between chromatic phenomena in stable homotopy theory and certain natural subrings of singular cohomology. This exploits the theory due to Henn, Lannes and Schwartz of unstable algebras over the Steenrod algebra localized away from nilpotents. △ Less

Submitted 17 April, 2009; originally announced April 2009.

Comments: This is the version published by Algebraic & Geometric Topology on 24 August 2006

MSC Class: 55N20; 55S10; 18F20

Journal ref: Algebr. Geom. Topol. 6 (2006) 1037-1067

Unstable modules over the Steenrod algebra revisited

Authors: Geoffrey M L Powell

Abstract: A new and natural description of the category of unstable modules over the Steenrod algebra as a category of comodules over a bialgebra is given; the theory extends and unifies the work of Carlsson, Kuhn, Lannes, Miller, Schwartz, Zarati and others. Related categories of comodules are studied, which shed light upon the structure of the category of unstable modules at odd primes. In particular, a… ▽ More A new and natural description of the category of unstable modules over the Steenrod algebra as a category of comodules over a bialgebra is given; the theory extends and unifies the work of Carlsson, Kuhn, Lannes, Miller, Schwartz, Zarati and others. Related categories of comodules are studied, which shed light upon the structure of the category of unstable modules at odd primes. In particular, a category of bigraded unstable modules is introduced; this is related to the study of modules over the motivic Steenrod algebra. △ Less

Submitted 28 March, 2009; originally announced March 2009.

Comments: This is the version published by Geometry & Topology Monographs on 14 November 2007

MSC Class: 55S10; 18E10

Journal ref: Geom. Topol. Monogr. 11 (2007) 245-288

Comparison of Characteristics and Practices amongst Spreadsheet Users with Different Levels of Experience

Authors: Kenneth R. Baker, Stephen G. Powell, Barry Lawson, Lynn Foster-Johnson

Abstract: We developed an internet-based questionnaire on spreadsheet use that we administered to a large number of users in several companies and organizations to document how spreadsheets are currently being developed and used in business. In this paper, we discuss the results drawn from of a comparison of responses from individuals with the most experience and expertise with those from individuals with… ▽ More We developed an internet-based questionnaire on spreadsheet use that we administered to a large number of users in several companies and organizations to document how spreadsheets are currently being developed and used in business. In this paper, we discuss the results drawn from of a comparison of responses from individuals with the most experience and expertise with those from individuals with the least. These results describe two views of spreadsheet design and use in organizations, and reflect gaps between these two groups and between these groups and the entire population of nearly 1600 respondents. Moreover, our results indicate that these gaps have multiple dimensions: they reflect not only the context, skill, and practices of individual users but also the policies of large organizations. △ Less

Submitted 2 March, 2008; originally announced March 2008.

Comments: 16 Pages, 11 Tables

ACM Class: J.1; H.4.1; K.6.4; D.2.5; D.2.9; K.8.1

Journal ref: Proc. European Spreadsheet Risks Int. Grp. (EuSpRIG) 2006 205-219 ISBN:1-905617-08-9

Impact of Errors in Operational Spreadsheets

Authors: Stephen G. Powell, Barry Lawson, Kenneth R. Baker

Abstract: All users of spreadsheets struggle with the problem of errors. Errors are thought to be prevalent in spreadsheets, and in some instances they have cost organizations millions of dollars. In a previous study of 50 operational spreadsheets we found errors in 0.8% to 1.8% of all formula cells, depending on how errors are defined. In the current study we estimate the quantitative impacts of errors i… ▽ More All users of spreadsheets struggle with the problem of errors. Errors are thought to be prevalent in spreadsheets, and in some instances they have cost organizations millions of dollars. In a previous study of 50 operational spreadsheets we found errors in 0.8% to 1.8% of all formula cells, depending on how errors are defined. In the current study we estimate the quantitative impacts of errors in 25 operational spreadsheets from five different organizations. We find that many errors have no quantitative impact on the spreadsheet. Those that have an impact often affect unimportant portions of the spreadsheet. The remaining errors do sometimes have substantial impacts on key aspects of the spreadsheet. This paper provides the first fully-documented evidence on the quantitative impact of errors in operational spreadsheets. △ Less

Submitted 4 January, 2008; originally announced January 2008.

Comments: 12 pages including references

ACM Class: J.1; H.4.1; K.6.4; D.2.5; D.2.9; K.8.1

Journal ref: Proc. European Spreadsheet Risks Int. Grp. 2007 57-68 ISBN 978-905617-58-6

Tunable dipolar magnetism in high-spin molecular clusters

Authors: M. Evangelisti, A. Candini, A. Ghirri, M. Affronte, G. W. Powell, I. A. Gass, P. A. Wood, S. Parsons, E. K. Brechin, D. Collison, S. L. Heath

Abstract: We report on the Fe17 high-spin molecular cluster and show that this system is an exemplification of nanostructured dipolar magnetism. Each Fe17 molecule, with spin S=35/2 and axial anisotropy as small as D=-0.02K, is the magnetic unit that can be chemically arranged in different packing crystals whilst preserving both spin ground-state and anisotropy. For every configuration, molecular spins ar… ▽ More We report on the Fe17 high-spin molecular cluster and show that this system is an exemplification of nanostructured dipolar magnetism. Each Fe17 molecule, with spin S=35/2 and axial anisotropy as small as D=-0.02K, is the magnetic unit that can be chemically arranged in different packing crystals whilst preserving both spin ground-state and anisotropy. For every configuration, molecular spins are correlated only by dipolar interactions. The ensuing interplay between dipolar energy and anisotropy gives rise to macroscopic behaviors ranging from superparamagnetism to long-range magnetic order at temperatures below 1K. △ Less

Submitted 20 September, 2006; v1 submitted 7 February, 2006; originally announced February 2006.

Comments: Replaced with version accepted for publication in Physical Review Letters

Journal ref: Phys. Rev. Lett. 97, 167202 (2006)