Dataset Viewer
text
stringlengths 2
350
| source
stringlengths 9
49
|
|---|---|
# 1
Prof. Antoine Bosselut
Modern Natural Language Processing - CS-552
09.04.2025 from 11 h 30 to 13 h 00
Duration : 90 minutes
## Midterm Practice Set (Solutions)
## SCIPER: 111111
Do not turn the page before the start of the exam. This document is double-sided, has 7 pages, the last ones are possibly blank. Do not unstaple. | mnlp_sample_exam.md_0_0 |
- This is a closed book exam. Non-programmable calculators are allowed. No other electronic devices of any kind are allowed.
- Place on your desk: your student ID, writing utensils, one double-sided A4 page cheat sheet if you have one; place all other personal items below your desk.
- You each have a different exam. | mnlp_sample_exam.md_0_1 |
- You each have a different exam.
- This exam has multiple-choice questions of varying difficulty. Each question is worth one point.
- Each question has exactly one correct answer. For each question, mark the box corresponding to the correct answer. You are not expected to get every question right even for the best grade. | mnlp_sample_exam.md_0_2 |
- Only answers in this booklet count. No extra loose answer sheets. You can use the blank pages at the end as scrap paper.
- Use a black or dark blue ballpen and clearly erase with correction fluid if necessary.
- If a question turns out to be wrong or ambiguous, we may decide to nullify it. | mnlp_sample_exam.md_0_3 |
Respectez les consignes suivantes : Observe this guidelines : Beachten Sie bitte die unten stehenden Richtlinien
choisir une réponse : select an answer $\mid$ ne PAS choisir une réponse : NOT select an answer $\mid$ Corriger une réponse : Correct an answer Antwort auswählen | mnlp_sample_exam.md_0_4 |
| $\times$ | $\times$ | $\times$ | $\times$ |
| :--: | :--: | :--: | :--: |
| | | | |
Question 1 Consider the equations for a standard LSTM cell: | mnlp_sample_exam.md_0_5 |
$$
\begin{aligned}
& i_{t}=\sigma\left(W_{i x} x_{t}+W_{i h} h_{t-1}+b_{i}\right) \\
& f_{t}=\sigma\left(W_{f x} x_{t}+W_{f h} h_{t-1}+b_{f}\right) \\
& o_{t}=\sigma\left(W_{o x} x_{t}+W_{o h} h_{t-1}+b_{o}\right) \\
& \tilde{c}_{t}=\phi\left(W_{c x} x_{t}+W_{c h} h_{t-1}+b_{c}\right) \\
& c_{t}=f_{t} \odot c_{t-1}+i_{t} \odot \tilde{c}_{t} \\ | mnlp_sample_exam.md_0_6 |
& h_{t}=o_{t} \odot \phi\left(c_{t}\right)
\end{aligned}
$$ | mnlp_sample_exam.md_0_7 |
In the equations above, which term explicitly represents the memory component that enables the LSTM to retain long-term information across timesteps? | mnlp_sample_exam.md_0_8 |
Output gate $o_{t}$
Hidden state $h_{t}$
Cell state $c_{t}$
Candidate cell state $\tilde{c}_{t}$
Input gate $i_{t}$
Solution: The cell state $c_{t}$ retains long-term information while the hidden state $h_{t}$ acts as a short-term memory. | mnlp_sample_exam.md_0_9 |
Question 2 BERT introduces a special token, [CLS], at the beginning of every input sequence. Which of the following statements best describes the purpose of the [CLS] token? | mnlp_sample_exam.md_0_10 |
It serves as a placeholder whose final hidden representation acts as a holistic sequence-level embedding, typically used for classification or next-sentence prediction tasks. | mnlp_sample_exam.md_0_11 |
It serves primarily to separate multiple sentences within the same input (the same role as [SEP] does).
$\square$ It simply marks sentence boundaries and carries no trainable embeddings of its own.
$\square$ It marks the exact midpoint of the input sequence to ensure balanced bidirectional attention. | mnlp_sample_exam.md_0_12 |
$\square$ It is used only during masked language modeling and is dropped for downstream tasks.
Solution: The [CLS] special token is introduced to aggregate information about the entire sequence in its embedding and is used as input to a classification model. | mnlp_sample_exam.md_0_13 |
Question 3 From the following set of models: \{ELMo, BERT, GPT, BART, T5\}, which group can each be directly used for both classification and generation tasks (without any modifications)? | mnlp_sample_exam.md_0_14 |
ELMo, BERT
$\square$ BERT, GPT
$\square$ BART, T5
$\square$ BERT, GPT, T5
ELMo, BART, GPT | mnlp_sample_exam.md_0_15 |
Solution: BART and T5 are encoder-decoder models capable of both classification and text generation. GPT also supports both tasks; however, in this question, it is always paired with bidirectional models like BERT and ELMo, which are not suitable for generation.Question 4 Which of the following best defines semantics-encoding embeddings of words? | mnlp_sample_exam.md_0_16 |
$\square$ A learned transformation of one-hot vectors into fixed-length random projections that improves computational efficiency.
$\square$ A technique that clusters words based on their dictionary definitions, ensuring that words with similar meanings always have identical representations. | mnlp_sample_exam.md_0_17 |
$\square$ A representation of words as vectors, where the relative distance encodes semantic similarity based on co-occurrence patterns.
$\square$ A representation of words as sparse vectors in a high-dimensional space, ensuring that each word has a unique but unrelated position in the space. | mnlp_sample_exam.md_0_18 |
Solution: Semantics-encoding embeddings capture meaning by placing words with similar usage patterns close together in vector space, typically trained using co-occurrence statistics. | mnlp_sample_exam.md_0_19 |
Question 5 Which of the following is FALSE regarding autoregressive natural language generation?
$\square$ At each step during inference, the model predicts a probability distribution over the vocabulary space.
$\square$ The next generated token is the one with maximum probability as predicted by the model. | mnlp_sample_exam.md_0_20 |
$\square$ We train a model to maximize the likelihood of the next token given the preceding tokens.
$\square$ Beam Search is more likely to generate a more probable sequence than greedy argmax decoding. | mnlp_sample_exam.md_0_21 |
Solution: This represents greedy decoding which is just one of the ways in which text can be generated. The next generated token may not be the most probable one in general. | mnlp_sample_exam.md_0_22 |
Question 6 Which of the following best describes Chain-of-Thought (CoT) prompting in large language models?
$\square$ A reinforcement learning technique that optimizes a language model's response generation using reward signals based on coherence and logical correctness. | mnlp_sample_exam.md_0_23 |
$\square$ A prompting technique where multiple examples are given in the prompt to guide the model towards correct predictions through imitation learning.
$\square$ A method where the model is fine-tuned on logical reasoning tasks to improve its structured decisionmaking capabilities. | mnlp_sample_exam.md_0_24 |
$\square$ A reasoning-based prompting method that encourages the model to break down complex problems into intermediate reasoning steps before producing a final answer. | mnlp_sample_exam.md_0_25 |
Solution: Chain-of-Thought prompting explicitly encourages models to generate intermediate reasoning steps to arrive at better final answers.Question 7 Which of the following claims are NOT TRUE about the perplexity metric?
(a) Easy to implement.
(b) Using base-2, base-3 and base-e will lead to the same perplexity score. | mnlp_sample_exam.md_0_26 |
(c) Can be cheated by predicting low-frequency tokens.
(d) Can be very sensitive to high frequency tokens.
(e) The perplexity score can be 0 . | mnlp_sample_exam.md_0_27 |
A and B.
$\square$ B and E.
$\square$ C and E.
$\square$ A and E.
$\square$ B, D and E
$\square$ B and C.
$\square$ C, D, E
$\square$ B, C and D.
Solution: C,D,E are not true. | mnlp_sample_exam.md_0_28 |
- C: can be cheated by predicting high-frequency tokens;
- D: sensitive to low frequency tokens;
- E: The perplexity score is a exponential of probability, which cannot be 0 . | mnlp_sample_exam.md_0_29 |
Question 8 Applying N-gram language model, Fixed-window language model and a RNN language model on the same training dataset, which one will have the largest model size (i.e. greatest number of parameters)?
$\square$ Cannot be determined from the given information
$\square$ Fixed-window language model
$\square$ RNN
$\square$ N-gram language model | mnlp_sample_exam.md_0_30 |
$\square$ N-gram language model
Solution: The scales of the fixed-window language model and RNN model do not depend on the training dataset. | mnlp_sample_exam.md_0_31 |
Figure 1: FFN Backpropagation
# Backpropagation
Question 9 According to Figure 1, the forward pass includes three steps:
$$
\begin{aligned}
& u=W_{\text {out }} h_{2}=w_{1}^{o} \times \phi_{12}(\cdot)+w_{2}^{o} \times \phi_{22}(\cdot)+w_{3}^{o} \times \phi_{32}(\cdot) \\
& \hat{y}=\phi_{0}(u) \\
& L(\hat{y}, y)=-y \ln \hat{y}
\end{aligned}
$$ | mnlp_sample_exam.md_0_32 |
Given that $\phi_{0}$ is the ReLU function, $W_{\text {out }}=\left[\begin{array}{lll}0.53 & -0.21 & 1.04\end{array}\right], h_{2}=\left[\begin{array}{lll}0.36 & 1.02 & 5.18\end{array}\right]^{T}, y=4.20$, what is the partial gradient $\frac{\partial L(\hat{y}, y)}{\partial \phi_{32}(\cdot)}$ ?
$\square 4.06$
$\square 0.81$
$\square 0.41$ | mnlp_sample_exam.md_0_33 |
$\square 0.81$
$\square 0.41$
$\square-4.06$
$\square-0.81$
$\square-0.41$
Solution: Computed by chain rule.
Question 10 Now given that the loss function is the L2 loss $L(\hat{y}, y)=\frac{1}{2}(y-\hat{y})^{2}$, what is the partial gradient $\frac{\partial L(\hat{y}, y)}{\partial w_{3}^{o}}$ ? | mnlp_sample_exam.md_0_34 |
1.21
$\square 0.42$
$\square 6.03$
$\square-1.21$
$\square-0.42$
$\square-6.03$
Solution: Computed by chain rule. | mnlp_sample_exam.md_0_35 |
[Jump to content](https://en.wikipedia.org/wiki/Machine_learning#bodyContent)
Main menu
Main menu
move to sidebar hide
Navigation
* [Main page](https://en.wikipedia.org/wiki/Main_Page "Visit the main page \[z\]")
* [Contents](https://en.wikipedia.org/wiki/Wikipedia:Contents "Guides to browsing Wikipedia") | ml.md_0_0 |
* [Current events](https://en.wikipedia.org/wiki/Portal:Current_events "Articles related to current events")
* [Random article](https://en.wikipedia.org/wiki/Special:Random "Visit a randomly selected article \[x\]")
* [About Wikipedia](https://en.wikipedia.org/wiki/Wikipedia:About "Learn about Wikipedia and how it works") | ml.md_0_1 |
* [Contact us](https://en.wikipedia.org/wiki/Wikipedia:Contact_us "How to contact Wikipedia") | ml.md_0_2 |
Contribute
* [Help](https://en.wikipedia.org/wiki/Help:Contents "Guidance on how to use and edit Wikipedia")
* [Learn to edit](https://en.wikipedia.org/wiki/Help:Introduction "Learn how to edit Wikipedia")
* [Community portal](https://en.wikipedia.org/wiki/Wikipedia:Community_portal "The hub for editors") | ml.md_0_3 |
* [Recent changes](https://en.wikipedia.org/wiki/Special:RecentChanges "A list of recent changes to Wikipedia \[r\]")
* [Upload file](https://en.wikipedia.org/wiki/Wikipedia:File_upload_wizard "Add images or other media for use on Wikipedia")
* [Special pages](https://en.wikipedia.org/wiki/Special:SpecialPages) | ml.md_0_4 |
[    ](https://en.wikipedia.org/wiki/Main_Page) | ml.md_0_5 |
[ Search ](https://en.wikipedia.org/wiki/Special:Search "Search Wikipedia \[f\]")
Search
Appearance
* [Donate](https://donate.wikimedia.org/?wmf_source=donate&wmf_medium=sidebar&wmf_campaign=en.wikipedia.org&uselang=en) | ml.md_0_6 |
* [Create account](https://en.wikipedia.org/w/index.php?title=Special:CreateAccount&returnto=Machine+learning "You are encouraged to create an account and log in; however, it is not mandatory") | ml.md_0_7 |
* [Log in](https://en.wikipedia.org/w/index.php?title=Special:UserLogin&returnto=Machine+learning "You're encouraged to log in; however, it's not mandatory. \[o\]") | ml.md_0_8 |
Personal tools
* [Donate](https://donate.wikimedia.org/?wmf_source=donate&wmf_medium=sidebar&wmf_campaign=en.wikipedia.org&uselang=en)
* [ Create account](https://en.wikipedia.org/w/index.php?title=Special:CreateAccount&returnto=Machine+learning "You are encouraged to create an account and log in; however, it is not mandatory") | ml.md_0_9 |
* [ Log in](https://en.wikipedia.org/w/index.php?title=Special:UserLogin&returnto=Machine+learning "You're encouraged to log in; however, it's not mandatory. \[o\]") | ml.md_0_10 |
Pages for logged out editors [learn more](https://en.wikipedia.org/wiki/Help:Introduction)
* [Contributions](https://en.wikipedia.org/wiki/Special:MyContributions "A list of edits made from this IP address \[y\]")
* [Talk](https://en.wikipedia.org/wiki/Special:MyTalk "Discussion about edits from this IP address \[n\]") | ml.md_0_11 |
## Contents
move to sidebar hide
* [ (Top) ](https://en.wikipedia.org/wiki/Machine_learning)
* [ 1 History ](https://en.wikipedia.org/wiki/Machine_learning#History)
* [ 2 Relationships to other fields ](https://en.wikipedia.org/wiki/Machine_learning#Relationships_to_other_fields) Toggle Relationships to other fields subsection | ml.md_0_12 |
* [ 2.1 Artificial intelligence ](https://en.wikipedia.org/wiki/Machine_learning#Artificial_intelligence)
* [ 2.2 Data compression ](https://en.wikipedia.org/wiki/Machine_learning#Data_compression)
* [ 2.3 Data mining ](https://en.wikipedia.org/wiki/Machine_learning#Data_mining) | ml.md_0_13 |
* [ 2.4 Generalization ](https://en.wikipedia.org/wiki/Machine_learning#Generalization)
* [ 2.5 Statistics ](https://en.wikipedia.org/wiki/Machine_learning#Statistics)
* [ 2.6 Statistical physics ](https://en.wikipedia.org/wiki/Machine_learning#Statistical_physics)
* [ 3 Theory ](https://en.wikipedia.org/wiki/Machine_learning#Theory) | ml.md_0_14 |
* [ 4 Approaches ](https://en.wikipedia.org/wiki/Machine_learning#Approaches) Toggle Approaches subsection
* [ 4.1 Supervised learning ](https://en.wikipedia.org/wiki/Machine_learning#Supervised_learning)
* [ 4.2 Unsupervised learning ](https://en.wikipedia.org/wiki/Machine_learning#Unsupervised_learning) | ml.md_0_15 |
* [ 4.3 Semi-supervised learning ](https://en.wikipedia.org/wiki/Machine_learning#Semi-supervised_learning)
* [ 4.4 Reinforcement learning ](https://en.wikipedia.org/wiki/Machine_learning#Reinforcement_learning)
* [ 4.5 Dimensionality reduction ](https://en.wikipedia.org/wiki/Machine_learning#Dimensionality_reduction) | ml.md_0_16 |
* [ 4.6 Other types ](https://en.wikipedia.org/wiki/Machine_learning#Other_types)
* [ 4.6.1 Self-learning ](https://en.wikipedia.org/wiki/Machine_learning#Self-learning)
* [ 4.6.2 Feature learning ](https://en.wikipedia.org/wiki/Machine_learning#Feature_learning) | ml.md_0_17 |
* [ 4.6.3 Sparse dictionary learning ](https://en.wikipedia.org/wiki/Machine_learning#Sparse_dictionary_learning)
* [ 4.6.4 Anomaly detection ](https://en.wikipedia.org/wiki/Machine_learning#Anomaly_detection)
* [ 4.6.5 Robot learning ](https://en.wikipedia.org/wiki/Machine_learning#Robot_learning) | ml.md_0_18 |
* [ 4.6.6 Association rules ](https://en.wikipedia.org/wiki/Machine_learning#Association_rules)
* [ 5 Models ](https://en.wikipedia.org/wiki/Machine_learning#Models) Toggle Models subsection
* [ 5.1 Artificial neural networks ](https://en.wikipedia.org/wiki/Machine_learning#Artificial_neural_networks) | ml.md_0_19 |
* [ 5.2 Decision trees ](https://en.wikipedia.org/wiki/Machine_learning#Decision_trees)
* [ 5.3 Random forest regression ](https://en.wikipedia.org/wiki/Machine_learning#Random_forest_regression)
* [ 5.4 Support-vector machines ](https://en.wikipedia.org/wiki/Machine_learning#Support-vector_machines) | ml.md_0_20 |
* [ 5.5 Regression analysis ](https://en.wikipedia.org/wiki/Machine_learning#Regression_analysis)
* [ 5.6 Bayesian networks ](https://en.wikipedia.org/wiki/Machine_learning#Bayesian_networks)
* [ 5.7 Gaussian processes ](https://en.wikipedia.org/wiki/Machine_learning#Gaussian_processes) | ml.md_0_21 |
* [ 5.8 Genetic algorithms ](https://en.wikipedia.org/wiki/Machine_learning#Genetic_algorithms)
* [ 5.9 Belief functions ](https://en.wikipedia.org/wiki/Machine_learning#Belief_functions)
* [ 5.10 Rule-based models ](https://en.wikipedia.org/wiki/Machine_learning#Rule-based_models) | ml.md_0_22 |
* [ 5.11 Training models ](https://en.wikipedia.org/wiki/Machine_learning#Training_models)
* [ 5.11.1 Federated learning ](https://en.wikipedia.org/wiki/Machine_learning#Federated_learning)
* [ 6 Applications ](https://en.wikipedia.org/wiki/Machine_learning#Applications) | ml.md_0_23 |
* [ 7 Limitations ](https://en.wikipedia.org/wiki/Machine_learning#Limitations) Toggle Limitations subsection
* [ 7.1 Explainability ](https://en.wikipedia.org/wiki/Machine_learning#Explainability)
* [ 7.2 Overfitting ](https://en.wikipedia.org/wiki/Machine_learning#Overfitting) | ml.md_0_24 |
* [ 7.3 Other limitations and vulnerabilities ](https://en.wikipedia.org/wiki/Machine_learning#Other_limitations_and_vulnerabilities)
* [ 8 Model assessments ](https://en.wikipedia.org/wiki/Machine_learning#Model_assessments)
* [ 9 Ethics ](https://en.wikipedia.org/wiki/Machine_learning#Ethics) Toggle Ethics subsection | ml.md_0_25 |
* [ 9.1 Bias ](https://en.wikipedia.org/wiki/Machine_learning#Bias)
* [ 9.2 Financial incentives ](https://en.wikipedia.org/wiki/Machine_learning#Financial_incentives)
* [ 10 Hardware ](https://en.wikipedia.org/wiki/Machine_learning#Hardware) Toggle Hardware subsection | ml.md_0_26 |
* [ 10.1 Tensor Processing Units (TPUs) ](https://en.wikipedia.org/wiki/Machine_learning#Tensor_Processing_Units_\(TPUs\))
* [ 10.2 Neuromorphic computing ](https://en.wikipedia.org/wiki/Machine_learning#Neuromorphic_computing) | ml.md_0_27 |
* [ 10.2.1 physical neural networks ](https://en.wikipedia.org/wiki/Machine_learning#physical_neural_networks)
* [ 10.3 Embedded machine learning ](https://en.wikipedia.org/wiki/Machine_learning#Embedded_machine_learning)
* [ 11 Software ](https://en.wikipedia.org/wiki/Machine_learning#Software) Toggle Software subsection | ml.md_0_28 |
* [ 11.1 Free and open-source software ](https://en.wikipedia.org/wiki/Machine_learning#Free_and_open-source_software)
* [ 11.2 Proprietary software with free and open-source editions ](https://en.wikipedia.org/wiki/Machine_learning#Proprietary_software_with_free_and_open-source_editions) | ml.md_0_29 |
* [ 11.3 Proprietary software ](https://en.wikipedia.org/wiki/Machine_learning#Proprietary_software)
* [ 12 Journals ](https://en.wikipedia.org/wiki/Machine_learning#Journals)
* [ 13 Conferences ](https://en.wikipedia.org/wiki/Machine_learning#Conferences)
* [ 14 See also ](https://en.wikipedia.org/wiki/Machine_learning#See_also) | ml.md_0_30 |
* [ 15 References ](https://en.wikipedia.org/wiki/Machine_learning#References)
* [ 16 Sources ](https://en.wikipedia.org/wiki/Machine_learning#Sources)
* [ 17 Further reading ](https://en.wikipedia.org/wiki/Machine_learning#Further_reading)
* [ 18 External links ](https://en.wikipedia.org/wiki/Machine_learning#External_links) | ml.md_0_31 |
Toggle the table of contents
# Machine learning
88 languages
* [Afrikaans](https://af.wikipedia.org/wiki/Masjienleer "Masjienleer â Afrikaans")
* [اÙعربÙØ©](https://ar.wikipedia.org/wiki/%D8%AA%D8%B9%D9%84%D9%85_%D8%A7%D9%84%D8%A2%D9%84%D8%A9 "تعÙÙ
اÙØ¢ÙØ© â Arabic") | ml.md_0_32 |
* [à¦
সমà§à¦¯à¦¼à¦¾](https://as.wikipedia.org/wiki/%E0%A6%AF%E0%A6%A8%E0%A7%8D%E0%A6%A4%E0%A7%8D%E0%A7%B0_%E0%A6%B6%E0%A6%BF%E0%A6%95%E0%A7%8D%E0%A6%B7%E0%A6%A3 "যনà§à¦¤à§à§° শিà¦à§à¦·à¦£ â Assamese") | ml.md_0_33 |
* [AzÉrbaycanca](https://az.wikipedia.org/wiki/Ma%C5%9F%C4%B1n_%C3%B6yr%C9%99nm%C9%99si "MaÅın öyrÉnmÉsi â Azerbaijani")
* [تÛرکجÙ](https://azb.wikipedia.org/wiki/%D9%85%D8%A7%D8%B4%DB%8C%D9%86_%D8%A7%D8%A4%DB%8C%D8%B1%D9%86%D9%85%D9%87%E2%80%8C%D8%B3%DB%8C "Ù
اشÛ٠اؤÛرÙÙ
ÙâØ³Û â South Azerbaijani") | ml.md_0_34 |
* [বাà¦à¦²à¦¾](https://bn.wikipedia.org/wiki/%E0%A6%AE%E0%A7%87%E0%A6%B6%E0%A6%BF%E0%A6%A8_%E0%A6%B2%E0%A6%BE%E0%A6%B0%E0%A7%8D%E0%A6%A8%E0%A6%BF%E0%A6%82 "মà§à¦¶à¦¿à¦¨ লারà§à¦¨à¦¿à¦ â Bangla")
* [é©åèª / Bân-lâm-gú](https://zh-min-nan.wikipedia.org/wiki/Ki-h%C4%81i_ha%CC%8Dk-si%CC%8Dp "Ki-hÄi haÌk-siÌp â Minnan") | ml.md_0_35 |
* [ÐаÑҡоÑÑÑа](https://ba.wikipedia.org/wiki/%D0%9C%D0%B0%D1%88%D0%B8%D0%BD%D0%B0%D0%BB%D1%8B_%D3%A9%D0%B9%D1%80%D3%99%D0%BD%D0%B5%D2%AF "ÐаÑÐ¸Ð½Ð°Ð»Ñ Ó©Ð¹ÑÓнеү â Bashkir") | ml.md_0_36 |
* [ÐелаÑÑÑкаÑ](https://be.wikipedia.org/wiki/%D0%9C%D0%B0%D1%88%D1%8B%D0%BD%D0%BD%D0%B0%D0%B5_%D0%BD%D0%B0%D0%B2%D1%83%D1%87%D0%B0%D0%BD%D0%BD%D0%B5 "ÐаÑÑннае навÑÑанне â Belarusian") | ml.md_0_37 |
* [à¤à¥à¤à¤ªà¥à¤°à¥](https://bh.wikipedia.org/wiki/%E0%A4%AE%E0%A4%B6%E0%A5%80%E0%A4%A8_%E0%A4%B2%E0%A4%B0%E0%A5%8D%E0%A4%A8%E0%A4%BF%E0%A4%82%E0%A4%97 "मशà¥à¤¨ लरà¥à¤¨à¤¿à¤à¤ â Bhojpuri") | ml.md_0_38 |
* [ÐÑлгаÑÑки](https://bg.wikipedia.org/wiki/%D0%9C%D0%B0%D1%88%D0%B8%D0%BD%D0%BD%D0%BE_%D1%81%D0%B0%D0%BC%D0%BE%D0%BE%D0%B1%D1%83%D1%87%D0%B5%D0%BD%D0%B8%D0%B5 "ÐаÑинно ÑамообÑÑение â Bulgarian") | ml.md_0_39 |
* [à½à½¼à½à¼à½¡à½²à½](https://bo.wikipedia.org/wiki/%E0%BD%A0%E0%BD%95%E0%BE%B2%E0%BD%B4%E0%BD%A3%E0%BC%8B%E0%BD%86%E0%BD%A6%E0%BC%8B%E0%BD%A6%E0%BE%B3%E0%BD%BC%E0%BD%96%E0%BC%8B%E0%BD%A6%E0%BE%A6%E0%BE%B1%E0%BD%BC%E0%BD%84%E0%BC%8B%E0%BC%8D "འà½à¾²à½´à½£à¼à½à½¦à¼à½¦à¾³à½¼à½à¼à½¦à¾¦à¾±à½¼à½à¼à¼ â Tibetan") | ml.md_0_40 |
* [Bosanski](https://bs.wikipedia.org/wiki/Ma%C5%A1insko_u%C4%8Denje "MaÅ¡insko uÄenje â Bosnian")
* [Català ](https://ca.wikipedia.org/wiki/Aprenentatge_autom%C3%A0tic "Aprenentatge automà tic â Catalan")
* [ÄeÅ¡tina](https://cs.wikipedia.org/wiki/Strojov%C3%A9_u%C4%8Den%C3%AD "Strojové uÄenà â Czech") | ml.md_0_41 |
* [Cymraeg](https://cy.wikipedia.org/wiki/Dysgu_peirianyddol "Dysgu peirianyddol â Welsh")
* [Dansk](https://da.wikipedia.org/wiki/Maskinl%C3%A6ring "Maskinlæring â Danish")
* [اÙدارجة](https://ary.wikipedia.org/wiki/%D8%AA%D8%B9%D9%84%D8%A7%D9%85_%D9%85%D8%A7%D9%83%D9%8A%D9%86%D9%8A "تعÙاÙ
Ù
اÙÙÙÙ â Moroccan Arabic") | ml.md_0_42 |
* [Deutsch](https://de.wikipedia.org/wiki/Maschinelles_Lernen "Maschinelles Lernen â German")
* [Eesti](https://et.wikipedia.org/wiki/Masin%C3%B5ppimine "Masinõppimine â Estonian") | ml.md_0_43 |
* [Îλληνικά](https://el.wikipedia.org/wiki/%CE%9C%CE%B7%CF%87%CE%B1%CE%BD%CE%B9%CE%BA%CE%AE_%CE%BC%CE%AC%CE%B8%CE%B7%CF%83%CE%B7 "ÎηÏανική μάθηÏη â Greek")
* [Español](https://es.wikipedia.org/wiki/Aprendizaje_autom%C3%A1tico "Aprendizaje automático â Spanish") | ml.md_0_44 |
* [Esperanto](https://eo.wikipedia.org/wiki/Ma%C5%9Dina_lernado "MaÅina lernado â Esperanto")
* [Euskara](https://eu.wikipedia.org/wiki/Ikasketa_automatiko "Ikasketa automatiko â Basque") | ml.md_0_45 |
* [ÙارسÛ](https://fa.wikipedia.org/wiki/%DB%8C%D8%A7%D8%AF%DA%AF%DB%8C%D8%B1%DB%8C_%D9%85%D8%A7%D8%B4%DB%8C%D9%86 "ÛادگÛØ±Û Ù
اشÛÙ â Persian")
* [Français](https://fr.wikipedia.org/wiki/Apprentissage_automatique "Apprentissage automatique â French") | ml.md_0_46 |
* [Gaelg](https://gv.wikipedia.org/wiki/Ynsaghey_jeshaghtagh "Ynsaghey jeshaghtagh â Manx")
* [Galego](https://gl.wikipedia.org/wiki/Aprendizaxe_autom%C3%A1tica "Aprendizaxe automática â Galician")
* [íêµì´](https://ko.wikipedia.org/wiki/%EA%B8%B0%EA%B3%84_%ED%95%99%EC%8A%B5 "ê¸°ê³ íìµ â Korean") | ml.md_0_47 |
* [ÕÕ¡ÕµÕ¥ÖÕ¥Õ¶](https://hy.wikipedia.org/wiki/%D5%84%D5%A5%D6%84%D5%A5%D5%B6%D5%A1%D5%B5%D5%A1%D5%AF%D5%A1%D5%B6_%D5%B8%D6%82%D5%BD%D5%B8%D6%82%D6%81%D5%B8%D6%82%D5%B4 "ÕÕ¥ÖÕ¥Õ¶Õ¡ÕµÕ¡Õ¯Õ¡Õ¶ Õ¸ÖÕ½Õ¸ÖÖÕ¸ÖÕ´ â Armenian") | ml.md_0_48 |
* [हिनà¥à¤¦à¥](https://hi.wikipedia.org/wiki/%E0%A4%AF%E0%A4%82%E0%A4%A4%E0%A5%8D%E0%A4%B0_%E0%A4%B6%E0%A4%BF%E0%A4%95%E0%A5%8D%E0%A4%B7%E0%A4%A3 "यà¤à¤¤à¥à¤° शिà¤à¥à¤·à¤£ â Hindi")
* [Ido](https://io.wikipedia.org/wiki/Mashinolernado "Mashinolernado â Ido") | ml.md_0_49 |
* [Bahasa Indonesia](https://id.wikipedia.org/wiki/Pemelajaran_mesin "Pemelajaran mesin â Indonesian")
* [IsiZulu](https://zu.wikipedia.org/wiki/Ukufunda_kwenguxa "Ukufunda kwenguxa â Zulu")
* [Ãslenska](https://is.wikipedia.org/wiki/V%C3%A9lan%C3%A1m "Vélanám â Icelandic") | ml.md_0_50 |
* [Italiano](https://it.wikipedia.org/wiki/Apprendimento_automatico "Apprendimento automatico â Italian")
* [×¢×ר×ת](https://he.wikipedia.org/wiki/%D7%9C%D7%9E%D7%99%D7%93%D7%AA_%D7%9E%D7%9B%D7%95%D7%A0%D7%94 "××××ת ×××× × â Hebrew")
* [Jawa](https://jv.wikipedia.org/wiki/Machine_learning "Machine learning â Javanese") | ml.md_0_51 |
* [à²à²¨à³à²¨à²¡](https://kn.wikipedia.org/wiki/%E0%B2%AF%E0%B2%82%E0%B2%A4%E0%B3%8D%E0%B2%B0_%E0%B2%95%E0%B2%B2%E0%B2%BF%E0%B2%95%E0%B3%86 "ಯà²à²¤à³à²° à²à²²à²¿à²à³ â Kannada") | ml.md_0_52 |
* [á¥áá áá£áá](https://ka.wikipedia.org/wiki/%E1%83%9B%E1%83%90%E1%83%9C%E1%83%A5%E1%83%90%E1%83%9C%E1%83%A3%E1%83%A0%E1%83%98_%E1%83%A1%E1%83%AC%E1%83%90%E1%83%95%E1%83%9A%E1%83%94%E1%83%91%E1%83%90 "áááá¥ááá£á á á¡á¬áááááá â Georgian") | ml.md_0_53 |
* [ÐÑÑгÑзÑа](https://ky.wikipedia.org/wiki/%D0%9C%D0%B0%D1%88%D0%B8%D0%BD%D0%B0_%D2%AF%D0%B9%D1%80%D3%A9%D0%BD%D2%AF%D2%AF "ÐаÑина үйÑөнүү â Kyrgyz")
* [LatvieÅ¡u](https://lv.wikipedia.org/wiki/Ma%C5%A1%C4%ABnm%C4%81c%C4%AB%C5%A1an%C4%81s "MaÅ¡Ä«nmÄcÄ«Å¡anÄs â Latvian") | ml.md_0_54 |
* [Lietuvių](https://lt.wikipedia.org/wiki/Ma%C5%A1ininis_mokymasis "MaÅ¡ininis mokymasis â Lithuanian")
* [Ligure](https://lij.wikipedia.org/wiki/Imprendimento_automatico "Imprendimento automatico â Ligurian")
* [Magyar](https://hu.wikipedia.org/wiki/G%C3%A9pi_tanul%C3%A1s "Gépi tanulás â Hungarian") | ml.md_0_55 |
* [ÐакедонÑки](https://mk.wikipedia.org/wiki/%D0%9C%D0%B0%D1%88%D0%B8%D0%BD%D1%81%D0%BA%D0%BE_%D1%83%D1%87%D0%B5%D1%9A%D0%B5 "ÐаÑинÑко ÑÑеÑе â Macedonian") | ml.md_0_56 |
* [മലയാളà´](https://ml.wikipedia.org/wiki/%E0%B4%AF%E0%B4%A8%E0%B5%8D%E0%B4%A4%E0%B5%8D%E0%B4%B0%E0%B4%AA%E0%B4%A0%E0%B4%A8%E0%B4%82 "യനàµà´¤àµà´°à´ªà´ നഠâ Malayalam") | ml.md_0_57 |
* [मराठà¥](https://mr.wikipedia.org/wiki/%E0%A4%AF%E0%A4%82%E0%A4%A4%E0%A5%8D%E0%A4%B0_%E0%A4%B6%E0%A4%BF%E0%A4%95%E0%A5%8D%E0%A4%B7%E0%A4%A3 "यà¤à¤¤à¥à¤° शिà¤à¥à¤·à¤£ â Marathi")
* [Bahasa Melayu](https://ms.wikipedia.org/wiki/Pembelajaran_mesin "Pembelajaran mesin â Malay") | ml.md_0_58 |
* [Ðонгол](https://mn.wikipedia.org/wiki/%D0%9C%D0%B0%D1%88%D0%B8%D0%BD_%D1%81%D1%83%D1%80%D0%B3%D0%B0%D0%BB%D1%82 "ÐаÑин ÑÑÑÐ³Ð°Ð»Ñ â Mongolian")
* [Nederlands](https://nl.wikipedia.org/wiki/Machinaal_leren "Machinaal leren â Dutch") | ml.md_0_59 |
* [æ¥æ¬èª](https://ja.wikipedia.org/wiki/%E6%A9%9F%E6%A2%B0%E5%AD%A6%E7%BF%92 "æ©æ¢°å¦ç¿ â Japanese")
* [Norsk bokmÃ¥l](https://no.wikipedia.org/wiki/Maskinl%C3%A6ring "Maskinlæring â Norwegian BokmÃ¥l")
* [Norsk nynorsk](https://nn.wikipedia.org/wiki/Maskinl%C3%A6ring "Maskinlæring â Norwegian Nynorsk") | ml.md_0_60 |
* [Occitan](https://oc.wikipedia.org/wiki/Aprendissatge_automatic "Aprendissatge automatic â Occitan") | ml.md_0_61 |
* [à¬à¬¡à¬¼à¬¿à¬](https://or.wikipedia.org/wiki/%E0%AC%AE%E0%AD%87%E0%AC%B8%E0%AC%BF%E0%AC%A8_%E0%AC%B2%E0%AC%B0%E0%AD%8D%E0%AC%A3%E0%AD%8D%E0%AC%A3%E0%AC%BF%E0%AC%99%E0%AD%8D%E0%AC%97%E0%AD%8D "ମàସିନ ଲରàଣàଣିà¬àà¬à â Odia") | ml.md_0_62 |
* [OÊ»zbekcha / ÑзбекÑа](https://uz.wikipedia.org/wiki/Mashina_o%CA%BBrganuvi "Mashina oÊ»rganuvi â Uzbek")
* [ਪੰà¨à¨¾à¨¬à©](https://pa.wikipedia.org/wiki/%E0%A8%AE%E0%A8%B8%E0%A8%BC%E0%A9%80%E0%A8%A8_%E0%A8%B2%E0%A8%B0%E0%A8%A8%E0%A8%BF%E0%A9%B0%E0%A8%97 "ਮਸ਼à©à¨¨ ਲਰਨਿੰਠâ Punjabi") | ml.md_0_63 |
End of preview. Expand
in Data Studio
- Downloads last month
- 165