---
title: 'MiniRocket: Deterministic Time-Series Transform'
url: https://www.emergentmind.com/topics/minimally-random-convolutional-kernel-transform-minirocket
type: topic
---

# MiniRocket: Deterministic Time-Series Transform

MiniRocket, short for **Minimally Random Convolutional Kernel Transform**, is a time-series classification transform in the ROCKET family that replaces fully random convolutional kernels with an almost deterministic, highly structured kernel bank, retains **Proportion of Positive Values (PPV)** as the sole pooling statistic, and trains a simple linear classifier on the resulting feature vector. In its original formulation, MiniRocket is **up to 75 times faster** than ROCKET on larger datasets while maintaining **essentially the same accuracy**, and it can train and test a classifier on all **109 datasets from the UCR archive in less than 10 minutes** [2012.08791]. Standard implementations produce about **9,996** scalar features per series and have become a common baseline for fast time-series classification [2012.08791][2409.01115].

## 1. Position within the ROCKET family

ROCKET introduced the now-standard pipeline of applying many 1D convolutional kernels to a time series, extracting simple global statistics from each activation map, and fitting a linear classifier on the transformed features. MiniRocket is a reformulation of that pipeline: it preserves the emphasis on dilation, PPV pooling, and linear classification, but removes most of the randomness in kernel generation and feature extraction, thereby shifting the method from a large random-feature transform toward an almost deterministic transform [2012.08791][2409.01115].

The canonical contrast is that ROCKET uses many random kernels with random length, weights, bias, dilation, and padding, whereas MiniRocket fixes the kernel length to **9**, uses a fixed family of kernels with two-valued weights, derives biases from convolution outputs rather than from a random uniform distribution, and discards global max pooling in favor of PPV alone [2012.08791][2409.01115]. Later papers summarize the weight restriction as values in \(\{-1,2\}\), and some describe equivalent sign-flipped forms such as \(\{-2,1\}\) [2409.01115][2511.01572].

| Aspect | ROCKET | MiniRocket |
|---|---|---|
| Kernel construction | Fully random kernels | Fixed, engineered kernel set |
| Pooling | PPV + global max pooling | PPV only |
| Typical feature count | \(20{,}000\) for \(10{,}000\) kernels | About \(9{,}996\) features |

MiniRocket later became the basis for several descendants and reinterpretations. MultiRocket expands the set of pooling operators and input representations; SelF-Rocket adds wrapper-based selection over representations and pooling operators; HDC-MiniROCKET reinterprets MiniRocket in hyperdimensional-computing terms and augments it with explicit time encoding [2409.01115][2202.08055].

## 2. Kernel parameterization and feature map

For a univariate time series \(\mathbf{X} = \{x_1,x_2,\ldots,x_T\}\), MiniRocket applies dilated 1D convolutions and converts each resulting activation map into a scalar PPV feature. A standard convolutional response can be written as
\[
(X * k)_t = \sum_{i=0}^{L-1} X_{t+id}\,w_i,
\]
and the PPV statistic for an activation map \(Z=\{z_1,\dots,z_n\}\) is
\[
PPV(Z)=\frac{1}{n}\sum_{i=1}^{n}[z_i>0].
\]
In MiniRocket, the feature for kernel \(k\) with associated bias \(b_k\) is typically written as
\[
\phi_k(X)=PPV(Z-b_k),
\]
and the standard feature vector contains approximately \(9{,}996\) such features [2409.01115].

The original MiniRocket paper fixes the kernel length to **9** and uses a subset of **84 kernels** drawn from the family of two-valued length-9 kernels. In the original construction, each kernel has weights \(\alpha=-1\) and \(\beta=2\), with exactly three positions assigned \(\beta\) and the remaining six assigned \(\alpha\), yielding \(84\) kernels from the relevant subset of the \(2^9=512\) possible two-valued patterns [2012.08791]. Later summaries describe the same design more generally as a fixed set of **84 kernels** with values in \(\{-1,2\}\), expanded by dilations into about \(9{,}996\) kernel/dilation combinations [2409.01115].

Bias generation is central to the method’s “minimal randomness.” Instead of drawing biases from \(\mathcal{U}(-1,1)\), MiniRocket samples bias values from quantiles of convolution outputs. In the default version, for each kernel/dilation combination, the bias quantiles are taken from the convolution output on **one randomly chosen training series**; in the fully deterministic variant, the quantiles are computed from convolution outputs over the **entire training set** [2012.08791]. This detail means that MiniRocket is not strictly randomness-free in its default form, even though the kernel bank itself is fixed [2012.08791].

The same transform is used beyond univariate classification. In prognostic classification on NASA’s CMAPSS data, the input is modeled as a multivariate time series
\[
X=\{x_1,\ldots,x_m\}\in\mathbb{R}^{m\times l},
\]
and MiniRocket yields one PPV feature per kernel for each multivariate series [2204.04527]. The method is also used on multivariate event-log-derived sequences for predictive maintenance and on raw or precomputed multichannel sensor data in human activity recognition [2305.10059][2402.18296].

## 3. Determinism, computational structure, and efficiency

MiniRocket retains the same broad asymptotic transform complexity as ROCKET,
\[
O(k\cdot n\cdot l_{\text{input}}),
\]
but drastically reduces constant factors through kernel regularization, PPV-only features, and shared computation across kernels [2012.08791]. The original implementation exploits the two-valued kernel structure by precomputing \(A=-X\) and \(G=3X\), constructing a shared convolution component for the \(\alpha\)-weighted positions, and adding the contributions of the three \(\beta\)-weighted positions for each kernel; the effect is that most work is shared across the **84** kernels for a fixed dilation [2012.08791].

The reported speed gains are large. On the **109 UCR datasets**, ROCKET required about **2 hours** of total compute time on a single CPU core, including about **1h 55m** for the transform, whereas MiniRocket required about **8 minutes** total, including about **2.5 minutes** for the transform [2012.08791]. On larger datasets such as **MosquitoSound**, **FruitFlies**, and **InsectSound**, MiniRocket was reported to be **43–75× faster** than ROCKET [2012.08791]. The original paper therefore characterizes MiniRocket as significantly faster than any other method of comparable accuracy, and significantly more accurate than methods of roughly similar computational expense [2012.08791].

The “almost deterministic” formulation also has a fully deterministic variant. Replacing per-kernel bias sampling from a random training series with bias estimation from the full training set yields a fully deterministic transform with empirically **essentially identical accuracy**, but at higher computational and memory cost [2012.08791]. In practice, the default MiniRocket configuration remains the usual choice.

Downstream classification is deliberately simple. The original MiniRocket work uses ridge regression on smaller datasets and logistic regression with Adam on larger datasets [2012.08791]. Later implementations and extensions commonly use **RidgeClassifier** as the default linear model, preserving the separation between a fixed convolutional transform and a lightweight classifier [2409.01115].

## 4. Accuracy profile and benchmark behavior

On the **109 UCR datasets**, MiniRocket is reported as **marginally more accurate** on average than ROCKET, and only slightly less accurate than methods such as **TS-CHIEF** and **HIVE-COTE/TDE**, with the differences to those stronger ensembles reported as **not statistically significant** under the paper’s pairwise tests [2012.08791]. The core result is therefore not merely faster inference, but a favorable **accuracy–cost** trade-off [2012.08791].

Later work clarified that MiniRocket’s strong average performance does not make PPV-only pooling universally optimal. A systematic study of **112 UCR datasets** constructed **15 MiniRocket variants** by combining five pooling operators with three input-representation sets and found that **PPV\_MIX** had the best average mean rank, yet it was **not** best on most individual datasets and was outperformed by another pooling/operator combination in about **78.57%** of datasets [2409.01115]. This result motivated SelF-Rocket, which keeps MiniRocket’s kernel design but dynamically selects the best representation and pooling operator during training; on the same benchmark, SelF-Rocket ranked **2nd overall**, below HIVE-COTE v2.0, and outperformed MiniRocket by about **1.23 percentage points** in mean accuracy difference [2409.01115].

A different line of work exposed a structural limitation of MiniRocket’s global PPV pooling. HDC-MiniROCKET showed that MiniRocket can suffer **catastrophic failures** when class identity depends primarily on global temporal position. On a synthetic two-class dataset of noisy signals containing a single sharp peak, with class determined only by whether the peak lay in the first or second half of the series, standard MiniROCKET reached about **65.0%** accuracy, while HDC-MiniROCKET with explicit time encoding reached **97.0%**; on a harder subset, MiniROCKET dropped to about **56.8%**, while HDC-MiniROCKET reached **94.1%** [2202.08055]. On **128 UCR datasets**, HDC-MiniROCKET improved over MiniROCKET on **81** datasets when the time-encoding scale parameter \(s\) was chosen by oracle search, with an average gain of about **3.1%** on those improved datasets [2202.08055].

These results collectively suggest a stable core conclusion. MiniRocket is a very strong default transform for general time-series classification, but its PPV-only, globally pooled representation can be suboptimal when either temporal localization or representation choice is critical [2202.08055][2409.01115].

## 5. Applications across domains

MiniRocket has been used well beyond the classical UCR benchmark. In prognostic health management on NASA’s **CMAPSS** turbofan dataset, MiniRocket was applied to health-status classification from multi-sensor time series. On **FD001** with **500 kernels**, MiniROCKET achieved **47.56%** accuracy, **38.17%** macro-F1, and **5 s** computation time, compared with **46.26%**, **41.13%**, and **23 s** for ROCKET, and much larger runtimes for stacked **LSTM** and **GRU** baselines. In a second experiment using **5000 kernels** on **FD001–FD004**, MiniROCKET remained much faster than the deep baselines, with a reported computation time of **25 s**, versus **111 s** for ROCKET and about **460–526 s** for the deep models [2204.04527]. The same study also showed that MiniRocket/ROCKET features can be combined with **SVM** and **LDA**, and on the difficult **FD004** subset, **SVM** and **LDA** on the random-kernel feature space reached **96.25% / 87.98%** and **95.34% / 90.70%** in accuracy/F1, respectively [2204.04527].

In predictive maintenance from ATM event logs, MiniROCKET was used after converting logs into multivariate time series of cumulative command–response counts in **10-minute** windows. With a Ridge classifier, MiniROCKET achieved average test **Accuracy \(0.7286\)**, **Balanced Accuracy \(0.6639\)**, **F1 \(0.3113\)**, **AUC \(0.6639\)**, **MS \(0.5814\)**, and **23.3 s** training time per fold [2305.10059]. The same study reported that MiniROCKET significantly outperformed **InceptionTime** on **Balanced Accuracy, F1, AUC, and MS**, while remaining statistically similar to ROCKET and much faster [2305.10059].

In smartphone-based human activity recognition on the UCI HAR dataset, MiniRocket was evaluated both on the standard **561-dimensional** handcrafted feature vectors and directly on raw single-channel sensor signals. On the precomputed-feature setup, MiniRocket achieved **Accuracy \(0.9881 \pm 0.0031\)**, **F1 \(0.9886 \pm 0.0029\)**, **AUC \(0.9932 \pm 0.0018\)**, with a training time of **80.1 seconds**; in raw single-channel mode, the best result came from the **Y-axis total acceleration** signal, with **Accuracy \(0.9350 \pm 0.0054\)**, **F1 \(0.9388 \pm 0.0051\)**, and **AUC \(0.9633 \pm 0.0031\)** [2402.18296].

MiniRocket has also been used successfully in EEG motor-imagery classification. On the PhysioNet MI-EEG dataset, a MiniRocket-plus-ridge-classifier pipeline achieved mean accuracy **98.63%**, compared with **98.06%** for a CNN–LSTM baseline, and the study explicitly states that classification via MiniRocket’s features achieved higher performance than the best deep learning models at lower computational cost [2508.16179].

In hyperspectral spectral classification, MiniROCKET was used as a fixed feature extractor on per-pixel spectra. A 2025 study reported that MiniROCKET **outperforms 1D-Justo-LiuNet in limited data scenarios and is mostly on par with it in the general case**, with the strongest advantage below roughly **100,000 samples per class**; on the **HYPSO-1** dataset with full training data, MiniROCKET achieved **OA 74.79**, compared with **72.90** for 1D-Justo-LiuNet [2509.13809].

## 6. Limitations, misconceptions, and ongoing directions

A common misconception is that “minimally random” means “fully deterministic.” In the original method, the kernel family, dilations, padding schedule, and feature type are fixed, but the default algorithm still samples bias quantiles from the convolution output of a **randomly chosen training series** for each kernel/dilation pair; full determinism requires a more expensive bias-estimation procedure over the full training set [2012.08791].

A second misconception is that PPV-only pooling is universally sufficient. The evidence is mixed. The original paper found that PPV alone matched or slightly exceeded PPV plus max pooling, which justified dropping global max pooling [2012.08791]. However, subsequent work showed that **pooling operators and input representations matter**, and that MiniRocket’s global PPV can suppress information about where patterns occur in the series, producing failures on tasks in which global temporal position is decisive [2409.01115][2202.08055].

Classifier choice is another important practical issue. MiniRocket is often presented as a transform plus ridge classifier, but the PHM literature showed that the same feature space can benefit from **SVM** or **LDA**, particularly in very high-dimensional settings and on difficult subsets such as **FD004** [2204.04527]. This suggests that MiniRocket itself should be separated conceptually from any single downstream linear model.

Interpretability remains limited. The feature space is high-dimensional, individual PPV features act as generic pattern detectors, and application studies note the absence of explicit feature selection or dimensionality reduction in standard usage [2204.04527]. In the hyperspectral setting, the fixed bias-fitting procedure was also identified as a possible source of suboptimal generalization because it relies only on the **first batch** for bias fitting in the reported implementation [2509.13809].

Several active directions therefore extend rather than replace MiniRocket. SelF-Rocket introduces dataset-specific selection over pooling operators and input representations [2409.01115]. HDC-MiniROCKET adds explicit time encoding while preserving MiniRocket’s convolutional front end [2202.08055]. Other papers note the possibility of richer input representations, filter-based feature selection, and kernel-pruning approaches such as **S-ROCKET**, **POCKET**, and **Detach-ROCKET** [2409.01115]. The general pattern is that MiniRocket remains the canonical fast baseline, while later work modifies its pooling, representation, temporal encoding, or feature-selection layers to address the cases in which PPV-only minimal randomness is not enough [2409.01115][2202.08055].

Source: https://www.emergentmind.com/topics/minimally-random-convolutional-kernel-transform-minirocket