Search

A composable infrastructure is defined as resources, such as compute, storage, accelerators and networking, that are shared in a pool and that can be grouped in various configurations to meet application requirements. This freedom to 'mix and match' resources dynamically allows for experimentation early in the design cycle, prior to the final architectural design or hardware implementation of a system. This design provides flexibility to serve a variety of workloads and provides a dynamic co-design platform that allows experiments and measurements in a controlled manner. For instance, key performance bottlenecks can be revealed early on in the experimentation phase thus avoiding costly and time consuming mistakes. Additionally, various system-level topologies can be evaluated when experimenting with new System on Chip (SoCs) and new accelerator types. This paper details the design of an enterprise composable infrastructure that we have implemented and made available to our partners in the IBM Research AI Hardware Center (AIHC). Our experimental evaluations on the composable system give insights into how the system works and evaluates the impact of various resource aggregations and reconfigurations on representative deep learning benchmarks., Comment: Submitted to IPDPS ScaDL 2021

Named Entity Recognition (NER) is the task of identifying and classifying named entities in unstructured text. In the legal domain, named entities of interest may include the case parties, judges, names of courts, case numbers, references to laws etc. We study the problem of legal NER with noisy text extracted from PDF files of filed court cases from US courts. The "gold standard" training data for NER systems provide annotation for each token of the text with the corresponding entity or non-entity label. We work with only partially complete training data, which differ from the gold standard NER data in that the exact location of the entities in the text is unknown and the entities may contain typos and/or OCR mistakes. To overcome the challenges of our noisy training data, e.g. text extraction errors and/or typos and unknown label indices, we formulate the NER task as a text-to-text sequence generation task and train a pointer generator network to generate the entities in the document rather than label them. We show that the pointer generator can be effective for NER in the absence of gold standard data and outperforms the common NER neural network architectures in long legal documents.

For Type Ia supernovae (SNe Ia) observed through a non-uniform interstellar medium (ISM) in its host galaxy, we investigate whether the non-uniformity can cause observable time variations in dust extinction and in gas absorption due to the expansion of the SN photosphere with time. We show that, owing to the steep spectral index of the ISM density power spectrum, sizable density fluctuation amplitudes at the length scale of typical ISM structures ($\gtrsim \text{ 10 pc}$) will translate to much smaller fluctuations on the scales of a SN photosphere. Therefore the typical amplitude of time variation due to non-uniform ISM, of absorption equivalent widths and of extinction, would be small. As a result, we conclude that non-uniform ISM density should not impact cosmology measurements based on SNe Ia. We apply our predictions based on the ISM density power law power spectrum to the observations of two highly reddened SNe Ia, SN 2012cu and SN 2014J., Comment: 13 pages, 1 figure

We present angle-resolved photoemission spectroscopy results for thin films of the three-dimensional manganese perovskite La$_{1-x}$Sr$_{x}$MnO$_{3}$. We show that the transition temperature ($T_c$) from the paramagnetic insulating to ferromagnetic metallic state is closely related to details of the electronic structure, particularly to the spectral weight at the ${\bf k}$-point, where the sharpest step at the Fermi level was observed. We found that this ${\bf k}$-point is the same for all the samples, despite their different $T_c$. The change of $T_c$ is discussed in terms of kinetic energy optimization. Our ARPES results suggest that the change of the electronic structure for the samples having different transition temperatures is different from the rigid band shift., Comment: Accepted by Journal of Physics: Condensed Matter

We construct supersymmetric deformations of general, locally supersymmetric, nonlinear sigma models in three spacetime dimensions, by extending the pure supergravity theory with a Chern-Simons term and gauging a subgroup of the sigma model isometries, possibly augmented with R-symmetry transformations. This class of models is shown to include theories with standard Yang-Mills Lagrangians, with optional moment interactions and topological mass terms. The results constitute a general classification of three-dimensional gauged supergravities., Comment: LaTeX2e, 51 pages

We give a pedagogical introduction to certain aspects of supersymmetric field theories in anti-de Sitter space. Among them are the presence of masslike terms in massless wave equations, irreducible unitary representations and the phenomenon of multiplet shortening., Comment: Lectures presented by B. de Wit at the Winter School of Theoretical Physics, Polanica, Poland, February 1999. 23 pp., LateX file, requires packages latexsym, amsfonts, cl2emult.cls

The rationale for using multi-model ensembles in climate change projections and impacts research is often based on the expectation that different models constitute independent estimates; therefore, a range of models allows a better characterisation of the uncertainties in the representation of the climate system than a single model. However, it is known that research groups share literature, ideas for representations of processes, parameterisations, evaluation data sets and even sections of model code. Thus, nominally different models might have similar biases because of similarities in the way they represent a subset of processes, or even be near-duplicates of others, weakening the assumption that they constitute independent estimates. If there are near-replicates of some models, then treating all models equally is likely to bias the inferences made using these ensembles. The challenge is to establish the degree to which this might be true for any given application. While this issue is recognised by many in the community, quantifying and accounting for model dependence in anything other than an ad-hoc way is challenging. Here we present a synthesis of the range of disparate attempts to define, quantify and address model dependence in multi-model climate ensembles in a common conceptual framework, and provide guidance on how users can test the efficacy of approaches that move beyond the equally weighted ensemble. In the upcoming Coupled Model Intercomparison Project phase 6 (CMIP6), several new models that are closely related to existing models are anticipated, as well as large ensembles from some models. We argue that quantitatively accounting for dependence in addition to model performance, and thoroughly testing the effectiveness of the approach used will be key to a sound interpretation of the CMIP ensembles in future scientific studies.