diff --git a/doc/doxyfile b/doc/doxyfile
index bcd11f1..9de24cd 100644
--- a/doc/doxyfile
+++ b/doc/doxyfile
@@ -650,7 +650,7 @@ HTML_FILE_EXTENSION = .html
# each generated HTML page. If it is left blank doxygen will generate a
# standard header.
-HTML_HEADER =
+HTML_HEADER = header.html
# The HTML_FOOTER tag can be used to specify a personal HTML footer for
# each generated HTML page. If it is left blank doxygen will generate a
diff --git a/doc/header.html b/doc/header.html
new file mode 100644
index 0000000..05b135e
--- /dev/null
+++ b/doc/header.html
@@ -0,0 +1,13 @@
+
+
+
+
+
+
+
+
+libLBFGS: L-BFGS library written in C
+
+
+
+
diff --git a/include/lbfgs.h b/include/lbfgs.h
index 1faa4cf..d23fc21 100644
--- a/include/lbfgs.h
+++ b/include/lbfgs.h
@@ -419,7 +419,7 @@ void lbfgs_free(lbfgsfloatval_t *x);
/**
-@mainpage A library of Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS)
+@mainpage libLBFGS: a library of Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS)
@section intro Introduction
@@ -445,7 +445,7 @@ drastically for large-scaled problems.
Among the various ports of L-BFGS, this library provides several features:
- Optimization with L1-norm (Orthant-Wise Limited-memory Quasi-Newton
- (OW-LQN) method):
+ (OWL-QN) method):
In addition to standard minimization problems, the library can minimize
a function F(x) combined with L1-norm |x| of the variables,
{F(x) + C |x|}, where C is a constant scalar parameter. This feature is
@@ -483,7 +483,7 @@ This library is used by:
@section download Download
-- Source code
+- Source code
libLBFGS is distributed under the term of the
MIT license.
@@ -495,8 +495,8 @@ libLBFGS is distributed under the term of the
method (::LBFGS_LINESEARCH_MORETHUENTE) or backtracking algorithm
(::LBFGS_LINESEARCH_BACKTRACKING).
- Fixed a bug: the previous version did not compute psuedo-gradients
- properly in the line search routines for OW-LQN. This bug might quit
- an iteration process too early when the OW-LQN routine was activated
+ properly in the line search routines for OWL-QN. This bug might quit
+ an iteration process too early when the OWL-QN routine was activated
(0 < ::lbfgs_parameter_t::orthantwise_c).
- Configure script for POSIX environments.
- SSE/SSE2 optimizations with GCC.
@@ -549,7 +549,7 @@ The line search algorithms used in this implementation are described in:
ACM Transactions on Mathematical Software (TOMS), Vol. 20, No. 3,
pp. 286-307, 1994.
-This library also implements Orthant-Wise Limited-memory Quasi-Newton (OW-LQN)
+This library also implements Orthant-Wise Limited-memory Quasi-Newton (OWL-QN)
method presented in:
- Galen Andrew and Jianfeng Gao.
Scalable training of L1-regularized log-linear models.