Notation: throughout, sums run over the samples in a single leaf, and n is the number of samples in that leaf.
1. The reported gains cannot be attributed to the proposed mechanism, and some results are impossible.
(a) By the paper's own figure, the shrinkage is numerically negligible. Fig. 4 (Jasmine, 10% noise, λ=1) shows all individual weights W≳0.9991, with cohort averages of about 0.99993 (clean) vs. 0.99991 (noisy) for label noise. Inverting W=1/(1+λσ2) gives λσ2≲9×10−4 at worst and ≈8×10−5 typically. From Eq. (5), every leaf value is plain GBDT's leaf value times a leaf-level factor:
cURB=cGBDT⋅1+λσˉ21,
where σˉ2 is the average of σ2 over the leaf. So leaves differ from GBDT by at most about 0.09%.
For split selection, each child's score (∑r)2/∑(1+λσ2) is GBDT's score (∑r)2/n times a factor in [1−δ,1], with δ≈9×10−4. Hence a split preferred by GBDT can be overturned only if its score exceeds the alternative's by less than about 0.09% (relative), i.e., only near-exact ties can change. This cannot produce the systematic gains reported, e.g., +22 points over GBDT on Madelon (0.9569 vs. 0.7327) and +11 over TabPFN on Bioresponse (0.9147 vs. 0.8019) (Table 1).
(b) Table 4 confirms this. At λ=0.1 the penalty is about 10−5, so URBoost is numerically GBDT. Yet it reaches 0.9364 accuracy on Jasmine with 10% noise, versus 0.8083 for GBDT on clean Jasmine (Table 1). Results are also unchanged across a 1000× range of λ (0.1–100). Both facts indicate the gains come from something other than variance regularization: a different base configuration, implementation, or evaluation pipeline.
Also URBoost's own configuration (number of trees, depth, learning rate) is never reported ("Official recommended parameters" do not exist for a new method).
The decisive λ=0 ablation under URBoost's configuration is missing as well.
(c) Some noise results violate basic information limits. If each training label is flipped with probability ρ, then
P(y~=1∣x)=(1−ρ)p(x)+ρ(1−p(x)),
which equals 0.5 at ρ=0.5: the observed labels are independent of the features. The paper's own baselines confirm this regime, dropping to about 0.5 at 50% noise in Fig. 2. Yet Table 4 reports Jasmine accuracy 0.9658 and AUC 0.9944 at 50% label noise. Accuracy also rises with noise: 0.9126 clean (Table 1), 0.9378 at 10%, 0.9658 at 50%. Fig. 2(a) appears to show the same on Madelon.
Separately, Jasmine baselines collapse to about 0.5 under 50% label noise (Fig. 2b) but stay at about 0.80–0.84 under 50% compound (label + feature) noise (Fig. 3b), although compound noise should be at least as harmful. These point to an error in the noise-injection or evaluation pipeline, so the robustness claims (4.2) cannot currently be trusted.
(d) Reporting overstates the results. Appendix F claims URBoost "consistently outperforms all competitors across every dataset and metric," but Table 2 reports only five of ten datasets. The five omitted include all three where URBoost is beaten in Table 1, even by untuned baselines:
- Elevators log loss: URBoost 0.2875 ranks 5th of 8 (TabPFN 0.2171, XGB 0.2619, LGBM 0.2732, CAT 0.2828).
- Letter: AUC 0.9988 ranks 6th; log loss 0.0134 ranks 5th; TabPFN is higher on accuracy and F1.
- Pol AUC: TabPFN 0.9998 vs. URBoost 0.9995.
In Table 1, baselines that beat URBoost are marked as significantly different. On Elevators log loss, TabPFN is marked at p<0.01 and XGB, LGBM and CAT at p<0.05; on Letter log loss, TabPFN is marked at p<0.01. The significance markers therefore cannot be read as "URBoost is significantly better," as the text implies.
2. URBoost's mechanism is a known regularizer in a new form, and its "instability" signal mostly measures how large a sample's residual is, not whether its label is wrong.
(a) URBoost is XGBoost's L2 leaf penalty, with the penalty strength set by trajectory variance. The paper does not mention this connection, but it follows directly from Eq. (5).
In XGBoost, each leaf outputs w=−G/(H+Λ). Here G and H are the sums of the first and second derivatives of the loss over the samples in the leaf, and Λ is the L2 penalty (reg_lambda). For squared error, each sample's first derivative is −r and its second derivative is 1. The leaf value therefore becomes ∑r/(n+Λ): the average residual, pulled slightly toward zero.
URBoost's leaf value (Eq. 5) is ∑r/(n+λ∑σ2). This is the same formula, with the fixed constant Λ replaced by λ∑σ2. Expanding URBoost's objective for a single leaf confirms it:
∑[(r−w)2+λσ2w2]=const+2[Gw+21(H+λ∑σ2)w2].
The right-hand side is XGBoost's leaf objective, up to a factor of 2 and a constant, with that replacement. Because the objectives match, the split scores match too, not only the leaf values. Put another way, URBoost is XGBoost with no L2 penalty, where each sample's second derivative is raised from 1 to 1+λσ2.
This has two consequences. First, the novelty is limited to making an existing penalty depend on trajectory variance. Second, the natural baselines are missing: XGBoost with tuned reg_lambda and min_child_weight, and URBoost implemented as an XGBoost custom objective (gradient −r, second derivative 1+λσ2). Appendix D does not tune reg_lambda at all.
How much does each method shrink a leaf? Compare each leaf's value with the plain average of its residuals, ∑r/n:
- XGBoost (default Λ=1):
- Under squared error, the leaf value is ∑r/(n+1), which is smaller than the plain average by a fraction 1/(n+1). That is 1% for a 100-sample leaf. Even the largest leaf possible on Jasmine (a 10-fold training set holds about 2,686 samples) is shrunk by 1/2687≈ 0.037%.
- Under log loss, XGBoost's shrinkage relative to its own unpenalized Newton step −G/H is 1/(H+1), where H=∑p(1−p). Since p(1−p)≤1/4, we have H≤n/4, so the shrinkage is even larger: at least 1/(2686/4+1)≈ 0.15% on Jasmine.
- URBoost: the leaf value is the plain average times 1/(1+λσˉ2), so it is shrunk by about λσˉ2. Fig. 4 puts this at roughly 0.008% typically and 0.09% at most.
So even XGBoost's weakest shrinkage is at least four to five times larger than URBoost's typical shrinkage, and about 19 times larger under log loss, which is what XGBoost uses on Jasmine. If pulling leaves toward zero were what made URBoost accurate, default XGBoost should do at least as well. Instead, it scores 0.80 on Jasmine against URBoost's 0.91 (Table 1). URBoost does aim its shrinkage at leaves with unstable samples rather than shrinking every leaf evenly, but even its strongest shrinkage (about 0.09%) is far too small to plausibly explain an 11-point gap.
(b) Trajectory variance is high for clean samples too. Take one clean sample that the trees fit well under squared loss. Write r(k) for its residual when tree k is fit, and h(k) for that tree's output on it. Fitting well means h(k)=r(k). Each update with learning rate ν then shrinks the residual by a constant factor, r(k+1)=(1−ν)r(k), so the trees' outputs decay smoothly: h(k)=r(1)(1−ν)k−1.
This is exactly the "stable descent" the paper associates with clean data. Yet the sequence still has nonzero variance. Taking the variance over k=1,…,K:
σ2=(r(1))2⋅Var[(1−ν)k−1]≈0.066(r(1))2for ν=0.1 and K=20 trees.
So a sample's trajectory variance grows with the square of its starting residual, whether or not its label is correct. Samples that start with large residuals include mislabeled ones, but also clean samples that are simply hard to fit.
The conclusion concedes that both kinds can have high variance, but the paper never measures how often this happens. Fig. 4(b) fits this reading: the clean cohort's weights also fall steadily over iterations, and the clean–noisy gap is only about 2×10−5.