We provide all the test data and corresponding predictions for our paper, “Practical Fluorescence Reconstruction Microscopy for High-Content Imaging”. Please refer to the Methods section in this paper for experimental details. For each experimental condition, we provide the input transmitted-light images (either phase contrast or DIC), the ground truth fluorescence images, and the output predicted fluorescence images which should reconstruct the ground truth fluorescence images.
Mondal, Shanka Subhra; Webb, Taylor; Cohen, Jonathan
Abstract:
A dataset of Raven’s Progressive Matrices (RPM)-like problems using realistically rendered
3D shapes, based on source code from CLEVR (a popular visual-question-answering dataset) (Johnson, J., Hariharan, B., Van Der Maaten, L., Fei-Fei, L., Lawrence Zitnick, C., & Girshick, R. (2017). Clevr: A diagnostic dataset for compositional language and elementary visual reasoning. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 2901-2910)).
Extrapolation -- the ability to make inferences that go beyond the scope of one's experiences -- is a hallmark of human intelligence. By contrast, the generalization exhibited by contemporary neural network algorithms is largely limited to interpolation between data points in their training corpora. In this paper, we consider the challenge of learning representations that support extrapolation. We introduce a novel visual analogy benchmark that allows the graded evaluation of extrapolation as a function of distance from the convex domain defined by the training data. We also introduce a simple technique, context normalization, that encourages representations that emphasize the relations between objects. We find that this technique enables a significant improvement in the ability to extrapolate, considerably outperforming a number of competitive techniques.
This item provides access to all configurations of single-chain nanoparticles analyzed in the manuscript "Sequence Patterning, Morphology, and Dispersity in Single-Chain Nanoparticles: Insights from Simulation and Machine Learning" by Roshan A. Patel, Sophia Colmenares, and Michael A. Webb (DOI: 10.1021/acspolymersau.3c00007). The single-chain nanoparticles derive from 320 unique precursor chains that are distinguished by the fraction of linker beads that decorate a fixed-length polymer backbone and the distribution or blockiness of those linker beads. The data is provided in the form of serialized object using the `pickle' python module. The data was compiled using Python version 3.8.8 and Clang 10.0.0. The Python object loaded from the .pkl file is a nested list, with the first dimension having 7,680 entries for the 7,680 unique single-chain nanoparticles produced in the aforementioned paper. Each of those 7,680 entries is itself a list with 20 entries, representing the 20 different simulation snapshots of the given single-chain nanoparticle. Each of the 20 entries is another list with two entries, with the first being a numpy.ndarray containing the x,y,z coordinates of all the beads comprising the single-chain nanoparticle and the second being a numpy.ndarray with a numerical encoding to indicate whether the beads are backbone (indicated as '0') or linker beads (indicated as '1'). Altogether, this provides 153,600 configurations of single-chain nanoparticles.
Here we publish the data used in paper "Junming Huang, Gavin Cook, and Yu Xie, Large-scale Quantitative Evidence of Media Impact on Public Opinion toward China". This dataset include estimated sentiments on The New York Times on China in eight topics from 1970 to 2019, and a time series of public attitude aggregated from surveys on China.
A scintillator type fast ion loss detector measures the gyroradius and pitch angle distribution of superthermal ions escaping from a magnetically confined fusion plasma at a single location. Described here is a technique for optimizing the angular orientation of such a detector in an axisymmetric tokamak geometry in order to intercept losses over a useful and interesting ranges of pitch angle. The method consists of evaluating the detector acceptance as a function of the fast ion constants of motion, i.e. energy, canonical toroidal momentum, and magnetic moment. The detector acceptance can then be plotted in a plane of constant energy and compared with the relevant orbit class boundaries and fast ion source distributions. Knowledge of expected or interesting mechanisms of loss can further guide selection of the detector orientation. The example of a fast ion loss detector for the National Spherical Torus Experiment-Upgrade (NSTX-U) is considered.
