I'm fitting and evaluating a CatBoostRegressor and a XGBRegressor to the same regression problem. I tried matching their hyperparameters as closely as possible, yet I'm seeing something strange: catboost test error is monotonically decreasing! Why is that? NOTE that I switched off the fancy boosting_type and it's plain.

catboost learning curve: catboost

xgboost learning curve: xgboost

catboost get_all_params:

{'nan_mode': 'Min',
 'eval_metric': 'RMSE',
 'iterations': 1000,
 'sampling_frequency': 'PerTree',
 'fold_permutation_block': 0,
 'leaf_estimation_method': 'Newton',
 'boosting_type': 'Plain',
 'feature_border_type': 'GreedyLogSum',
 'bayesian_matrix_reg': 0.1000000015,
 'l2_leaf_reg': 3,
 'random_strength': 1,
 'rsm': 1,
 'boost_from_average': True,
 'model_size_reg': 0.5,
 'approx_on_full_history': False,
 'subsample': 0.8000000119,
 'use_best_model': True,
 'random_seed': 0,
 'depth': 6,
 'has_time': False,
 'fold_len_multiplier': 2,
 'border_count': 254,
 'classes_count': 0,
 'sparse_features_conflict_fraction': 0,
 'leaf_estimation_backtracking': 'AnyImprovement',
 'best_model_min_trees': 1,
 'model_shrink_rate': 0,
 'loss_function': 'RMSE',
 'learning_rate': 0.009999999776,
 'score_function': 'Cosine',
 'task_type': 'CPU',
 'leaf_estimation_iterations': 1,
 'bootstrap_type': 'MVS',
 'permutation_count': 4}

xgboost get_params:

{'base_score': 0.5,
 'booster': 'gbtree',
 'colsample_bylevel': 1,
 'colsample_bynode': 1,
 'colsample_bytree': 1,
 'gamma': 0,
 'importance_type': 'gain',
 'learning_rate': 0.01,
 'max_delta_step': 0,
 'max_depth': 6,
 'min_child_weight': 1,
 'missing': None,
 'n_estimators': 1000,
 'n_jobs': 1,
 'nthread': None,
 'objective': 'reg:linear',
 'random_state': 0,
 'reg_alpha': 0,
 'reg_lambda': 1,
 'scale_pos_weight': 1,
 'seed': None,
 'silent': False,
 'subsample': 1,
 'verbosity': 1}
  • 1
    $\begingroup$ In addition to the general principles outlined in the answer from @usεr11852 it seems that catboost is still learning (very slowly) at 700 trees. The minimum test MSE is about 1.2 for xgboost while the last test MSE for catbost seems to be about 1.4; see if going out to a few thousand trees finds a minimum closer to 1.2 for catboost. $\endgroup$
    – EdM
    Feb 25 '20 at 15:39
  • $\begingroup$ @EdM (+1) Yeah, obviously, I wanted to make that point too but forgot! xgboost achieves a substantially better minima way sooner. $\endgroup$
    – usεr11852
    Feb 25 '20 at 16:43

As it stands the two runs are not directly comparable. For example, off the bat, we can see that:

  1. catboost uses a stronger $L_2$ regularisation on the weights ($3$ instead of $1$ in xgboost).
  2. catboost uses different bagging/subsampling in terms of rows ($0.8$ instead of $1.0$ in xgboost).
  3. catboost uses a different way than xgboost to built its trees; symmetric instead of best-first in XGBoost.
  4. catboost uses an extra regularisation parameter (bayesian_matrix_reg) to regularise leaf values calculation (the non-diagonal ones when computing the hessian on the leafs); strictly speaking xgboost do not have the same functionality.

Please note that the above points are not even a very in-depth look. While we might expect the same qualitative behaviour between different implementations of a general computational framework (e.g. lowering the depth of the base-learner trees decreases the chance of over-fitting), the effect size might not be comparable. As these implementations incorporate a number of computational techniques to accelerate and/or make them more stable, the direct comparison of their behaviour is unreliable.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.