Figure 7 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Confusion matrix showing average and standard deviation over five runs for categorical classification on the balanced test set.
Source publication
We present a novel method of classifying Type Ia supernovae using convolutional neural networks, a neural network framework typically used for image recognition. Our model is trained on photometric information only, eliminating the need for accurate redshift data. Photometric data is pre-processed via 2D Gaussian process regression into two-dimensi...
Contexts in source publication
Context 1
... accuracy, and 98.18 ± 0.3% test accuracy. The confusion matrices in Figure 7 show the average and standard deviations of the by-type breakdown for 5 independent runs. It is worth noting that we trained and tested with the PLAsTiCC dataset even though it is not class-balanced for this task to try to evaluate the model's performance on a dataset emulating the relative frequencies of these events in nature. ...
Context 2
... data in Figures 7 and 8 show some level of symmetry between misclassifications. SNIax and SLSN-1 seem to be easily distinguishable across the board, for example, with 0's in all relevant cells in both figures except one. ...
Context 3
... and SLSN-1 seem to be easily distinguishable across the board, for example, with 0's in all relevant cells in both figures except one. In Figure 7, SNIbc and SLSN-1 are seemingly very similar, as they are misclassified as one another at similarly high rates. ...
Context 4
... are notable differences between Figures 7 and 8, however. In Figure 7, representing the classifier's performance on class-balanced categorical classification, the model mispredicts SNIa's as other types at a similar rate as non-Ia's mispredicted as Ia's. An average of 3.2 Ia's are mispredicted, whereas an average of 4 non-Ia's are misclassified as Ia. ...
Context 5
... misclassified set summarized in Table 7 is the result of one of the five class-balanced categorical classification runs represented by the data in Figure 7. ...
Context 6
... point of interest is the lack of symmetry between misclassifications, in contrast with the analysis of Fig- ures 7 and 8. This is clear in the significantly larger number of SLSN-1 misclassified as SNIbc (7) compared with the number of SNIbc misclassified as SLSN-1 (1). ...
Context 7
... is also more often misclassified as other types (3 as SNIa, 3 as SNIbc, and 1 as SNIa-91bg) than non-Iax misclassified as Iax (1 SNIa and 1 SNIbc). The more symmetric Figure 7 suggests that the asymmetry of this table is due to randomness and would be corrected with data from other runs. ...
Similar publications
In this work, we present classification results on early supernova light curves from SCONE, a photometric classifier that uses convolutional neural networks to categorize supernovae (SNe) by type using light-curve data. SCONE is able to identify SN types from light curves at any stage, from the night of initial alert to the end of their lifetimes....
In this work, we present classification results on early supernova lightcurves from SCONE, a photometric classifier that uses convolutional neural networks to categorize supernovae (SNe) by type using lightcurve data. SCONE is able to identify SN types from lightcurves at any stage, from the night of initial alert to the end of their lifetimes. Sim...
We put constraints on the peak absolute magnitude of type Ia supernova using the Pantheon sample for type Ia supernova observations and the cosmic chronometers data for the Hubble parameter by a model independent and non-parametric approach. Our analysis is based on the Gaussian process regression. We find percent level bounds on the peak absolute...
