You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
href="https://doi.org/10.1093/gigascience/giaa026">ShinyLearner: A
257
-
containerized benchmarking tool for machine-learning classification of
258
-
tabular data.</a>
252
+
<strong>Title</strong>: <ahref="https://doi.org/10.1093/gigascience/giaa026">ShinyLearner: A containerized benchmarking tool for machine-learning classification of tabular data.</a>
259
253
</p>
260
254
<p>
261
-
<strong>Authors</strong>: Terry J Lee, Erica Suh, Kimball Hill, Stephen
262
-
R Piccolo
255
+
<strong>Authors</strong>: Terry J Lee, Erica Suh, Kimball Hill, Stephen R Piccolo
software is a resource to researchers who wish to benchmark multiple
305
-
classification or feature-selection algorithms on a given dataset. We
306
-
hope it will serve as example of combining the benefits of software
307
-
containerization with a user-friendly approach.</jats:p></jats:sec>
264
+
<jats:title>Abstract</jats:title><jats:sec><jats:title>Background</jats:title><jats:p>Classification algorithms assign observations to groups based on patterns in data. The machine-learning community have developed myriad classification algorithms, which are used in diverse life science research domains. Algorithm choice can affect classification accuracy dramatically, so it is crucial that researchers optimize the choice of which algorithm(s) to apply in a given research domain on the basis of empirical evidence. In benchmark studies, multiple algorithms are applied to multiple datasets, and the researcher examines overall trends. In addition, the researcher may evaluate multiple hyperparameter combinations for each algorithm and use feature selection to reduce data dimensionality. Although software implementations of classification algorithms are widely available, robust benchmark comparisons are difficult to perform when researchers wish to compare algorithms that span multiple software packages. Programming interfaces, data formats, and evaluation procedures differ across software packages; and dependency conflicts may arise during installation.</jats:p></jats:sec><jats:sec><jats:title>Findings</jats:title><jats:p>To address these challenges, we created ShinyLearner, an open-source project for integrating machine-learning packages into software containers. ShinyLearner provides a uniform interface for performing classification, irrespective of the library that implements each algorithm, thus facilitating benchmark comparisons. In addition, ShinyLearner enables researchers to optimize hyperparameters and select features via nested cross-validation; it tracks all nested operations and generates output files that make these steps transparent. ShinyLearner includes a Web interface to help users more easily construct the commands necessary to perform benchmark comparisons. ShinyLearner is freely available at <ahref="https://github.com/srp33/ShinyLearner" class="uri">https://github.com/srp33/ShinyLearner</a>.</jats:p></jats:sec><jats:sec><jats:title>Conclusions</jats:title><jats:p>This software is a resource to researchers who wish to benchmark multiple classification or feature-selection algorithms on a given dataset. We hope it will serve as example of combining the benefits of software containerization with a user-friendly approach.</jats:p></jats:sec>
Only visualiation steps performed, rather than machine learning (which
348
-
could take several hours/days). The created figures match those in the
349
-
article. The content of other output files was not checked.
299
+
Only visualiation steps performed, rather than machine learning (which could take several hours/days). The created figures match those in the article. The content of other output files was not checked.
A neural net was used to analyse samples of natural images and text. For
272
-
the natural images, components resemble derivatives of Gaussian
273
-
operators, similar to those found in visual cortex and inferred from
274
-
psychophysics. While the results from natural images do not depend on
275
-
scale, those from text images are highly scale dependent. Convolution of
276
-
one of the text components with an original image shows that it is
277
-
sensitive to inter-word gaps.
264
+
A neural net was used to analyse samples of natural images and text. For the natural images, components resemble derivatives of Gaussian operators, similar to those found in visual cortex and inferred from psychophysics. While the results from natural images do not depend on scale, those from text images are highly scale dependent. Convolution of one of the text components with an original image shows that it is sensitive to inter-word gaps.
href="https://doi.org/10.1073/pnas.79.8.2554">Neural networks and
257
-
physical systems with emergent collective computational abilities</a>
252
+
<strong>Title</strong>: <ahref="https://doi.org/10.1073/pnas.79.8.2554">Neural networks and physical systems with emergent collective computational abilities</a>
<jats:p>Computational properties of use of biological organisms or to
275
-
the construction of computers can emerge as collective properties of
276
-
systems having a large number of simple equivalent components (or
277
-
neurons). The physical meaning of content-addressable memory is
278
-
described by an appropriate phase space flow of the state of a system. A
279
-
model of such a system is given, based on aspects of neurobiology but
280
-
readily adapted to integrated circuits. The collective properties of
281
-
this model produce a content-addressable memory which correctly yields
282
-
an entire memory from any subpart of sufficient size. The algorithm for
283
-
the time evolution of the state of the system is based on asynchronous
284
-
parallel processing. Additional emergent collective properties include
285
-
some capacity for generalization, familiarity recognition,
286
-
categorization, error correction, and time sequence retention. The
287
-
collective properties are only weakly sensitive to details of the
288
-
modeling or the failure of individual devices.</jats:p>
264
+
<jats:p>Computational properties of use of biological organisms or to the construction of computers can emerge as collective properties of systems having a large number of simple equivalent components (or neurons). The physical meaning of content-addressable memory is described by an appropriate phase space flow of the state of a system. A model of such a system is given, based on aspects of neurobiology but readily adapted to integrated circuits. The collective properties of this model produce a content-addressable memory which correctly yields an entire memory from any subpart of sufficient size. The algorithm for the time evolution of the state of the system is based on asynchronous parallel processing. Additional emergent collective properties include some capacity for generalization, familiarity recognition, categorization, error correction, and time sequence retention. The collective properties are only weakly sensitive to details of the modeling or the failure of individual devices.</jats:p>
0 commit comments