DOC: fix letter for hyper-parameter of IPCRidge

Closes #454

doc/user_guide/coxnet.ipynb

@@ -16,7 +16,7 @@
     "- \\frac{\\alpha}{2} \\sum_{j=1}^p \\beta_j^2 ,\n",
     "$$\n",
     "\n",
-    "where $\\mathrm{PL}(\\beta)$ is the partial likelihood function of the Cox model, $\\beta_1,\\ldots,\\beta_p$ are the coefficients for $p$ features, and $\\alpha \\geq 0$ is a hyper-parameter that controls the amount of shrinkage. The resulting objective is often referred to as *ridge regression*. If $\\lambda$ is set to zero, we obtain the standard, unpenalized Cox model."
+    "where $\\mathrm{PL}(\\beta)$ is the partial likelihood function of the Cox model, $\\beta_1,\\ldots,\\beta_p$ are the coefficients for $p$ features, and $\\alpha \\geq 0$ is a hyper-parameter that controls the amount of shrinkage. The resulting objective is often referred to as *ridge regression*. If $\\alpha$ is set to zero, we obtain the standard, unpenalized Cox model."
    ]
   },
   {