diff --git a/R-package/R/aliases.R b/R-package/R/aliases.R
index 5fcba4f46fe7..7cd3245727af 100644
--- a/R-package/R/aliases.R
+++ b/R-package/R/aliases.R
@@ -113,6 +113,7 @@
             , "num_trees"
             , "num_round"
             , "num_rounds"
+            , "nrounds"
             , "num_boost_round"
             , "n_estimators"
             , "max_iter"
diff --git a/R-package/tests/testthat/test_basic.R b/R-package/tests/testthat/test_basic.R
index 5aff8e2c2d2c..92b11650c3e2 100644
--- a/R-package/tests/testthat/test_basic.R
+++ b/R-package/tests/testthat/test_basic.R
@@ -188,6 +188,68 @@ test_that("lightgbm() rejects negative or 0 value passed to nrounds", {
   }
 })
 
+test_that("lightgbm() accepts nrounds as either a top-level argument or parameter", {
+  nrounds <- 15L
+
+  set.seed(708L)
+  top_level_bst <- lightgbm(
+    data = train$data
+    , label = train$label
+    , nrounds = nrounds
+    , params = list(
+      objective = "regression"
+      , metric = "l2"
+      , num_leaves = 5L
+    )
+    , save_name = tempfile(fileext = ".model")
+  )
+
+  set.seed(708L)
+  param_bst <- lightgbm(
+    data = train$data
+    , label = train$label
+    , params = list(
+      objective = "regression"
+      , metric = "l2"
+      , num_leaves = 5L
+      , nrounds = nrounds
+    )
+    , save_name = tempfile(fileext = ".model")
+  )
+
+  set.seed(708L)
+  both_customized <- lightgbm(
+    data = train$data
+    , label = train$label
+    , nrounds = 20L
+    , params = list(
+      objective = "regression"
+      , metric = "l2"
+      , num_leaves = 5L
+      , nrounds = nrounds
+    )
+    , save_name = tempfile(fileext = ".model")
+  )
+
+  top_level_l2 <- top_level_bst$eval_train()[[1L]][["value"]]
+  params_l2    <- param_bst$eval_train()[[1L]][["value"]]
+  both_l2      <- both_customized$eval_train()[[1L]][["value"]]
+
+  # check type just to be sure the subsetting didn't return a NULL
+  expect_true(is.numeric(top_level_l2))
+  expect_true(is.numeric(params_l2))
+  expect_true(is.numeric(both_l2))
+
+  # check that model produces identical performance
+  expect_identical(top_level_l2, params_l2)
+  expect_identical(both_l2, params_l2)
+
+  expect_identical(param_bst$current_iter(), top_level_bst$current_iter())
+  expect_identical(param_bst$current_iter(), both_customized$current_iter())
+  expect_identical(param_bst$current_iter(), nrounds)
+
+})
+
 test_that("lightgbm() performs evaluation on validation sets if they are provided", {
   set.seed(708L)
   dvalid1 <- lgb.Dataset(
@@ -467,6 +529,76 @@ test_that("lgb.train() rejects negative or 0 value passed to nrounds", {
   }
 })
 
+
+test_that("lgb.train() accepts nrounds as either a top-level argument or parameter", {
+  nrounds <- 15L
+
+  set.seed(708L)
+  top_level_bst <- lgb.train(
+    data = lgb.Dataset(
+      train$data
+      , label = train$label
+    )
+    , nrounds = nrounds
+    , params = list(
+      objective = "regression"
+      , metric = "l2"
+      , num_leaves = 5L
+      , save_name = tempfile(fileext = ".model")
+    )
+  )
+
+  set.seed(708L)
+  param_bst <- lgb.train(
+    data = lgb.Dataset(
+      train$data
+      , label = train$label
+    )
+    , params = list(
+      objective = "regression"
+      , metric = "l2"
+      , num_leaves = 5L
+      , nrounds = nrounds
+      , save_name = tempfile(fileext = ".model")
+    )
+  )
+
+  set.seed(708L)
+  both_customized <- lgb.train(
+    data = lgb.Dataset(
+      train$data
+      , label = train$label
+    )
+    , nrounds = 20L
+    , params = list(
+      objective = "regression"
+      , metric = "l2"
+      , num_leaves = 5L
+      , nrounds = nrounds
+      , save_name = tempfile(fileext = ".model")
+    )
+  )
+
+  top_level_l2 <- top_level_bst$eval_train()[[1L]][["value"]]
+  params_l2    <- param_bst$eval_train()[[1L]][["value"]]
+  both_l2      <- both_customized$eval_train()[[1L]][["value"]]
+
+  # check type just to be sure the subsetting didn't return a NULL
+  expect_true(is.numeric(top_level_l2))
+  expect_true(is.numeric(params_l2))
+  expect_true(is.numeric(both_l2))
+
+  # check that model produces identical performance
+  expect_identical(top_level_l2, params_l2)
+  expect_identical(both_l2, params_l2)
+
+  expect_identical(param_bst$current_iter(), top_level_bst$current_iter())
+  expect_identical(param_bst$current_iter(), both_customized$current_iter())
+  expect_identical(param_bst$current_iter(), nrounds)
+
+})
+
+
 test_that("lgb.train() throws an informative error if 'data' is not an lgb.Dataset", {
   bad_values <- list(
     4L
diff --git a/docs/Parameters.rst b/docs/Parameters.rst
index 088773d0690e..75bef7add9bc 100644
--- a/docs/Parameters.rst
+++ b/docs/Parameters.rst
@@ -153,7 +153,7 @@ Core Parameters
 
    -  **Note**: can be used only in CLI version
 
--  ``num_iterations`` :raw-html:`<a id="num_iterations" title="Permalink to this parameter" href="#num_iterations">&#x1F517;&#xFE0E;</a>`, default = ``100``, type = int, aliases: ``num_iteration``, ``n_iter``, ``num_tree``, ``num_trees``, ``num_round``, ``num_rounds``, ``num_boost_round``, ``n_estimators``, ``max_iter``, constraints: ``num_iterations >= 0``
+-  ``num_iterations`` :raw-html:`<a id="num_iterations" title="Permalink to this parameter" href="#num_iterations">&#x1F517;&#xFE0E;</a>`, default = ``100``, type = int, aliases: ``num_iteration``, ``n_iter``, ``num_tree``, ``num_trees``, ``num_round``, ``num_rounds``, ``nrounds``, ``num_boost_round``, ``n_estimators``, ``max_iter``, constraints: ``num_iterations >= 0``
 
    -  number of boosting iterations
 
diff --git a/include/LightGBM/config.h b/include/LightGBM/config.h
index c94420645cd9..da43a5ec9782 100644
--- a/include/LightGBM/config.h
+++ b/include/LightGBM/config.h
@@ -161,7 +161,7 @@ struct Config {
   // desc = **Note**: can be used only in CLI version
   std::vector<std::string> valid;
 
-  // alias = num_iteration, n_iter, num_tree, num_trees, num_round, num_rounds, num_boost_round, n_estimators, max_iter
+  // alias = num_iteration, n_iter, num_tree, num_trees, num_round, num_rounds, nrounds, num_boost_round, n_estimators, max_iter
   // check = >=0
   // desc = number of boosting iterations
   // desc = **Note**: internally, LightGBM constructs ``num_class * num_iterations`` trees for multi-class classification problems
diff --git a/python-package/lightgbm/basic.py b/python-package/lightgbm/basic.py
index 812fd82a5d97..569d9383f680 100644
--- a/python-package/lightgbm/basic.py
+++ b/python-package/lightgbm/basic.py
@@ -386,6 +386,7 @@ class _ConfigAliases:
                                   "num_trees",
                                   "num_round",
                                   "num_rounds",
+                                  "nrounds",
                                   "num_boost_round",
                                   "n_estimators",
                                   "max_iter"},
diff --git a/src/io/config_auto.cpp b/src/io/config_auto.cpp
index 4e3f000a88f5..cd24790b820c 100644
--- a/src/io/config_auto.cpp
+++ b/src/io/config_auto.cpp
@@ -33,6 +33,7 @@ const std::unordered_map<std::string, std::string>& Config::alias_table() {
   {"num_trees", "num_iterations"},
   {"num_round", "num_iterations"},
   {"num_rounds", "num_iterations"},
+  {"nrounds", "num_iterations"},
   {"num_boost_round", "num_iterations"},
   {"n_estimators", "num_iterations"},
   {"max_iter", "num_iterations"},