Google Professional Machine Learning Engineer Practice Test 62
Question 610
At a Google Cloud analytics team preparing a governed pilot, the following condition occurs: A retailer has no trustworthy segment labels but attempts a supervised classifier with synthetic classes supplied by the sales team. The team must satisfy a reproducible low-code AI assessment. What acceptance result verifies the intended behavior after remediation?
- An unseen labeled cohort receives probabilities for the intended positive class with a calibrated operating threshold.
- On held-out questions, answers remain grounded while input tokens and p95 response latency decline.
- A trained candidate reaches the agreed quality threshold without maintaining unsupported bespoke training infrastructure.
- Stable clusters reveal coherent customer behavior without pretending an unsupervised assignment is known supervised ground truth.
Explanation
Correct Answer
D:Stable clusters reveal coherent customer behavior without pretending an unsupervised assignment is known supervised ground truth. This is specific acceptance evidence for bigquery ml kmeans customer segmentation. K-means discovers unlabeled similarity groups; fabricating labels would make supervised accuracy meaningless. It confirms the controlled correction: Train an appropriately scaled BigQuery ML KMEANS model and assess cluster cohesion, size and business interpretability. Here, the relevant situation is: the following condition occurs: A retailer has no trustworthy segment labels but attempts a supervised classifier….
Incorrect Answers
A:An unseen labeled cohort receives probabilities for the intended positive class with a calibrated operating threshold. This result could demonstrate success for bigquery ml logistic model for conversion probability but not for the stated bigquery ml kmeans customer segmentation issue. The direct test is: Stable clusters reveal coherent customer behavior without pretending an unsupervised assignment is known supervised ground truth.
B:On held-out questions, answers remain grounded while input tokens and p95 response latency decline. This result could demonstrate success for gemini prompt context token budget but not for the stated bigquery ml kmeans customer segmentation issue. The direct test is: Stable clusters reveal coherent customer behavior without pretending an unsupervised assignment is known supervised ground truth.
C:A trained candidate reaches the agreed quality threshold without maintaining unsupported bespoke training infrastructure. This result could demonstrate success for automl tabular supervised model selection but not for the stated bigquery ml kmeans customer segmentation issue. The direct test is: Stable clusters reveal coherent customer behavior without pretending an unsupervised assignment is known supervised ground truth.
Question 611
During a regulated MLOps pipeline deployment, the following condition occurs: A Cloud Storage finalize event triggers training for every object, including temporary files, causing duplicate and premature model runs. The workflow must be repeatable and protect the last approved production model. Which underlying engineering principle explains why the original approach fails?
- Storage event notifications require business-level readiness and deduplication before training begins.
- Training success does not guarantee compatibility with the deployed model's consumer API.
- Kubernetes resource limits constrain component runtime even when the model's training job itself has spare capacity.
- Model retraining reproducibility requires stable inputs rather than unversioned reads of an evolving table.
Explanation
Correct Answer
A:Storage event notifications require business-level readiness and deduplication before training begins. This explains the mechanism behind eventbridge equivalent event trigger on storage and why the described failure is possible. A discriminating investigation is: Inspect object prefixes, event filters, completed-batch manifests and recent pipeline execution IDs. The corresponding repair is: Filter validated completed-dataset events and enforce an idempotent manifest-based training trigger. This matters here because a Cloud Storage finalize event triggers training for every object, including temporary files, causing duplicate and….
Incorrect Answers
B:Training success does not guarantee compatibility with the deployed model's consumer API. This principle describes continuous delivery integration contract tests, which is not the cause of the observed eventbridge equivalent event trigger on storage failure. The controlling explanation is: Storage event notifications require business-level readiness and deduplication before training begins.
C:Kubernetes resource limits constrain component runtime even when the model's training job itself has spare capacity. This principle describes custom kfp component resource limits, which is not the cause of the observed eventbridge equivalent event trigger on storage failure. The controlling explanation is: Storage event notifications require business-level readiness and deduplication before training begins.
D:Model retraining reproducibility requires stable inputs rather than unversioned reads of an evolving table. This principle describes data version reproducibility in ct jobs, which is not the cause of the observed eventbridge equivalent event trigger on storage failure. The controlling explanation is: Storage event notifications require business-level readiness and deduplication before training begins. This matters here because a Cloud Storage finalize event triggers training for every object, including temporary files, causing duplicate and….
Question 612
During a regulated MLOps pipeline deployment, the following condition occurs: A Cloud Storage finalize event triggers training for every object, including temporary files, causing duplicate and premature model runs. The workflow must be repeatable and protect the last approved production model. Select TWO: the decisive diagnostic step and the targeted repair for this specific issue.
- Create immutable versioned input snapshots or partition manifests and record them in training metadata.
- Inspect object prefixes, event filters, completed-batch manifests and recent pipeline execution IDs.
- Enforce a shared typed schema contract and reject incompatible feature versions before training or inference.
- Inspect label arrival times, cohort maturity, feedback collection coverage and effective training cutoffs.
- Filter validated completed-dataset events and enforce an idempotent manifest-based training trigger.
Explanation
Correct Answers
B:Inspect object prefixes, event filters, completed-batch manifests and recent pipeline execution IDs. This inspects evidence relevant to eventbridge equivalent event trigger on storage. Storage event notifications require business-level readiness and deduplication before training begins. After identifying the cause, the matching correction is: Filter validated completed-dataset events and enforce an idempotent manifest-based training trigger.
E:Filter validated completed-dataset events and enforce an idempotent manifest-based training trigger. This is the targeted corrective response for eventbridge equivalent event trigger on storage. Storage event notifications require business-level readiness and deduplication before training begins. Verify the diagnosis using: Inspect object prefixes, event filters, completed-batch manifests and recent pipeline execution IDs. Confirm success using: One completed batch starts one intended pipeline run and unrelated object uploads do not trigger training.
Incorrect Answers
A:Create immutable versioned input snapshots or partition manifests and record them in training metadata. This is an action or diagnostic relevant to data version reproducibility in ct jobs, not one of the required two steps. In this scenario verify: Inspect object prefixes, event filters, completed-batch manifests and recent pipeline execution IDs. Then correct: Filter validated completed-dataset events and enforce an idempotent manifest-based training trigger.
C:Enforce a shared typed schema contract and reject incompatible feature versions before training or inference. This is an action or diagnostic relevant to serving schema validation component, not one of the required two steps. In this scenario verify: Inspect object prefixes, event filters, completed-batch manifests and recent pipeline execution IDs. Then correct: Filter validated completed-dataset events and enforce an idempotent manifest-based training trigger.
D:Inspect label arrival times, cohort maturity, feedback collection coverage and effective training cutoffs. This is an action or diagnostic relevant to scheduled retraining on delayed ground truth, not one of the required two steps. In this scenario verify: Inspect object prefixes, event filters, completed-batch manifests and recent pipeline execution IDs. Then correct: Filter validated completed-dataset events and enforce an idempotent manifest-based training trigger.
Question 613
While preparing a production training workflow on Gemini Enterprise Agent Platform, the following condition occurs: Distributed training validation counts the final padded examples twice because the evaluation sampler duplicates data to balance worker batches. Engineers need an auditable, resource-aware correction to the issue. Which outcome most directly demonstrates a successful, measurable fix?
- Every evaluation feature is available at its simulated prediction time and later-period scores remain stable.
- No duplicated or synthetic neighbor crosses the evaluation boundary and rare-class recall reflects unseen defects.
- The job schedules successfully with the approved hardware and remains inside budget and residency constraints.
- The distributed evaluation matches a single-worker reference score and each real example contributes once.
Explanation
Correct Answer
D:The distributed evaluation matches a single-worker reference score and each real example contributes once. This is specific acceptance evidence for data-parallel evaluation duplicate samples. Distributed sampling can pad replicas; uncorrected metric aggregation may bias evaluation results. It confirms the controlled correction: Deduplicate or correctly weight validation samples when aggregating distributed metrics across workers. This matters here because while preparing a production training workflow on Gemini Enterprise Agent Platform, the following condition occurs:….
Incorrect Answers
A:Every evaluation feature is available at its simulated prediction time and later-period scores remain stable. This result could demonstrate success for temporal training label leakage in rolling windows but not for the stated data-parallel evaluation duplicate samples issue. The direct test is: The distributed evaluation matches a single-worker reference score and each real example contributes once.
B:No duplicated or synthetic neighbor crosses the evaluation boundary and rare-class recall reflects unseen defects. This result could demonstrate success for training class imbalance fold leakage but not for the stated data-parallel evaluation duplicate samples issue. The direct test is: The distributed evaluation matches a single-worker reference score and each real example contributes once. This matters here because while preparing a production training workflow on Gemini Enterprise Agent Platform, the following condition occurs:….
C:The job schedules successfully with the approved hardware and remains inside budget and residency constraints. This result could demonstrate success for accelerator quota and regional availability but not for the stated data-parallel evaluation duplicate samples issue. The direct test is: The distributed evaluation matches a single-worker reference score and each real example contributes once.
Question 614
During a staged restoration test at a secondary serving deployment, the following condition occurs: A new model revision passes deployment health checks but produces many application-level 5xx errors after traffic increases; rollback was never configured. The team must prove an actual user-visible improvement without weakening access controls. Which targeted engineering change most directly addresses the failure?
- Benchmark a Cloud Run custom container against a managed Agent Platform endpoint and choose for workload fit.
- Apply bounded concurrency, idempotent retries with exponential backoff and explicit overload responses.
- Pin releases to reviewed immutable image digests and record image plus model artifact versions.
- Add guarded traffic shifts and automatic rollback to the retained healthy revision when threshold alarms trip.
Explanation
Correct Answer
D:Add guarded traffic shifts and automatic rollback to the retained healthy revision when threshold alarms trip. This is the targeted corrective response for model traffic rollback on high error rate. Container readiness checks alone do not guarantee successful predictions under real traffic and data distributions. Verify the diagnosis using: Inspect endpoint revision metrics, release alarms, traffic fractions and availability of the prior known-good model. Confirm success using: A controlled failed rollout restores the prior model and returns errors below the service objective.
Incorrect Answers
A:Benchmark a Cloud Run custom container against a managed Agent Platform endpoint and choose for workload fit. This action addresses cloud run versus managed inference endpoint, not the failure in the described model traffic rollback on high error rate case. The targeted correction is: Add guarded traffic shifts and automatic rollback to the retained healthy revision when threshold alarms trip.
B:Apply bounded concurrency, idempotent retries with exponential backoff and explicit overload responses. This action addresses predictive serving rate limit and backpressure, not the failure in the described model traffic rollback on high error rate case. The targeted correction is: Add guarded traffic shifts and automatic rollback to the retained healthy revision when threshold alarms trip.
C:Pin releases to reviewed immutable image digests and record image plus model artifact versions. This action addresses model artifact image digest reproducibility, not the failure in the described model traffic rollback on high error rate case. The targeted correction is: Add guarded traffic shifts and automatic rollback to the retained healthy revision when threshold alarms trip.
Question 615
During a staged restoration test at a secondary serving deployment, the following condition occurs: A customer-facing inference API has high end-to-end latency in Asia while model execution is quick and the endpoint runs on another continent. The team must prove an actual user-visible improvement without weakening access controls. Which evidence most directly establishes what is wrong with this AI or ML workflow?
- Decompose client network, TLS, request transport, model compute and response times by user geography.
- Compare requested entity key bytes, stored feature-group schema, casting rules and online lookup failures.
- Compare request rate, queue depth, GPU utilization, per-instance concurrency and p95 response latency.
- Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata.
Explanation
Correct Answer
A:Decompose client network, TLS, request transport, model compute and response times by user geography. This inspects evidence relevant to serving region proximity versus model compute. Network distance can dominate inference response times when the actual model execution is already short. After identifying the cause, the matching correction is: Select supported regional placement or routing close to users while respecting residency and model availability. This matters here because during a staged restoration test at a secondary serving deployment, the following condition occurs: A….
Incorrect Answers
B:Compare requested entity key bytes, stored feature-group schema, casting rules and online lookup failures. This examines feature online entity key mismatch rather than the specific mechanism reported here. It is not the decisive first check for serving region proximity versus model compute. The relevant evidence is: Decompose client network, TLS, request transport, model compute and response times by user geography.
C:Compare request rate, queue depth, GPU utilization, per-instance concurrency and p95 response latency. This examines throughput-based autoscaling versus cpu metric rather than the specific mechanism reported here. It is not the decisive first check for serving region proximity versus model compute. The relevant evidence is: Decompose client network, TLS, request transport, model compute and response times by user geography.
D:Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata. This examines model registry version provenance rather than the specific mechanism reported here. It is not the decisive first check for serving region proximity versus model compute. The relevant evidence is: Decompose client network, TLS, request transport, model compute and response times by user geography.
Question 616
During a controlled replay of the data and model handoff to a second engineering team, the following condition occurs: Two experiments use the same model configuration but show different accuracy due to untracked data shuffling and nondeterministic seeds. The handoff must preserve privacy, lineage and testable behavior. Which evidence most directly establishes what is wrong with this AI or ML workflow?
- Inspect pipeline execution IDs, experiment tracking records, artifact URIs and the link between metrics and saved models.
- Inspect random seeds, sample ordering, framework determinism settings and environment versions recorded per run.
- Inspect join-key frequency distributions, per-task runtime, shuffle spill and the output cardinality before optimization.
- Inspect dataset and split fingerprints, label definitions, metric versions and evaluation timestamps for both experiment runs.
Explanation
Correct Answer
B:Inspect random seeds, sample ordering, framework determinism settings and environment versions recorded per run. This inspects evidence relevant to experiment random seed and data order variance. A single stochastic training run is not sufficient to distinguish algorithm improvement from sampling variation. After identifying the cause, the matching correction is: Log reproducibility settings and rerun comparable trials with fixed split and controlled randomness before attributing changes.
Incorrect Answers
A:Inspect pipeline execution IDs, experiment tracking records, artifact URIs and the link between metrics and saved models. This examines pipeline experiment versus notebook-only tracking rather than the specific mechanism reported here. It is not the decisive first check for experiment random seed and data order variance. The relevant evidence is: Inspect random seeds, sample ordering, framework determinism settings and environment versions recorded per run.
C:Inspect join-key frequency distributions, per-task runtime, shuffle spill and the output cardinality before optimization. This examines managed spark join-key skew repair rather than the specific mechanism reported here. It is not the decisive first check for experiment random seed and data order variance. The relevant evidence is: Inspect random seeds, sample ordering, framework determinism settings and environment versions recorded per run.
D:Inspect dataset and split fingerprints, label definitions, metric versions and evaluation timestamps for both experiment runs. This examines experiment metric comparison with mismatched splits rather than the specific mechanism reported here. It is not the decisive first check for experiment random seed and data order variance. The relevant evidence is: Inspect random seeds, sample ordering, framework determinism settings and environment versions recorded per run.
Question 617
During model-training validation on a Google Cloud experimental project, the following condition occurs: An eight-GPU job scales only slightly beyond two GPUs because each small batch requires excessive cross-worker gradient synchronization. The change should preserve a meaningful independent model-quality comparison. Which technical explanation best accounts for the reported behavior?
- Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated.
- TPU and GPU performance depends on operations, data pipeline and framework integration, not headline compute alone.
- Resampling before splitting leaks information and makes minority-class performance appear artificially high.
- Data parallel replicas each hold a model copy; they do not solve a model-too-large-for-one-device problem.
Explanation
Correct Answer
A:Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated. This explains the mechanism behind distributed gradient communication bottleneck and why the described failure is possible. A discriminating investigation is: Inspect all-reduce time, network bandwidth, effective batch size, worker stragglers and samples per second. The corresponding repair is: Tune supported distributed batch sizing, communication strategy or placement after measuring the bottleneck. This matters here because during model-training validation on a Google Cloud experimental project, the following condition occurs: An….
Incorrect Answers
B:TPU and GPU performance depends on operations, data pipeline and framework integration, not headline compute alone. This principle describes tpu versus gpu for tensorflow workload, which is not the cause of the observed distributed gradient communication bottleneck failure. The controlling explanation is: Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated.
C:Resampling before splitting leaks information and makes minority-class performance appear artificially high. This principle describes training class imbalance fold leakage, which is not the cause of the observed distributed gradient communication bottleneck failure. The controlling explanation is: Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated.
D:Data parallel replicas each hold a model copy; they do not solve a model-too-large-for-one-device problem. This principle describes model parallelism for memory-bound transformer, which is not the cause of the observed distributed gradient communication bottleneck failure. The controlling explanation is: Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated. This matters here because during model-training validation on a Google Cloud experimental project, the following condition occurs: An….
Question 618
During a production ML inference release on Google Cloud, the following condition occurs: A private inference service exists with proper invoker IAM, but a VPC client cannot connect because the private route and DNS mapping are missing. The service must meet its agreed quality and latency requirements. Which result is the strongest postchange evidence of success?
- The morning surge meets availability and p95 latency objectives while off-peak cost remains documented.
- Regional clients experience lower p95 end-to-end latency without changes to model compute quality.
- A test client on the approved network succeeds while an external probe remains blocked.
- Approved in-network clients can invoke predictions while unapproved public network paths cannot reach the service.
Explanation
Correct Answer
C:A test client on the approved network succeeds while an external probe remains blocked. This is specific acceptance evidence for private vpc routing failure for serving. Identity permission is not proof of network reachability; private serving also requires correct routing and name resolution. It confirms the controlled correction: Configure the supported private DNS, network route and endpoint access path without granting public ingress. This matters here because a private inference service exists with proper invoker IAM, but a VPC client cannot connect because the private….
Incorrect Answers
A:The morning surge meets availability and p95 latency objectives while off-peak cost remains documented. This result could demonstrate success for autoscaling minimum capacity versus cold load but not for the stated private vpc routing failure for serving issue. The direct test is: A test client on the approved network succeeds while an external probe remains blocked.
B:Regional clients experience lower p95 end-to-end latency without changes to model compute quality. This result could demonstrate success for serving region proximity versus model compute but not for the stated private vpc routing failure for serving issue. The direct test is: A test client on the approved network succeeds while an external probe remains blocked.
D:Approved in-network clients can invoke predictions while unapproved public network paths cannot reach the service. This result could demonstrate success for private endpoint for regulated inference traffic but not for the stated private vpc routing failure for serving issue. The direct test is: A test client on the approved network succeeds while an external probe remains blocked.
Question 619
During a postdeployment AI safety and reliability review, the following condition occurs: An LLM summarizer produces fluent answers containing unsupported policy details, but the team measures only response latency and vocabulary similarity. The model owner must provide repeatable evidence rather than a cosmetic dashboard change. Which technical explanation best accounts for the reported behavior?
- Predictive importance is not causal proof; attribution techniques describe model behavior under their assumptions.
- Fluent text or semantic similarity is not proof an answer is supported by the retrieved source material.
- Supervised production quality requires reliable prediction-to-ground-truth linkage, not only unlabeled model outputs.
- Excess prompt context can increase processing cost and latency without improving response accuracy.
Explanation
Correct Answer
B:Fluent text or semantic similarity is not proof an answer is supported by the retrieved source material. This explains the mechanism behind generative ai factual grounding evaluation and why the described failure is possible. A discriminating investigation is: Inspect retrieved evidence, claim-level factual consistency, reference answers and human review samples. The corresponding repair is: Add groundedness and factual correctness checks to a representative gen AI evaluation suite and tune retrieval or prompting accordingly.
Incorrect Answers
A:Predictive importance is not causal proof; attribution techniques describe model behavior under their assumptions. This principle describes explainability versus causal claim, which is not the cause of the observed generative ai factual grounding evaluation failure. The controlling explanation is: Fluent text or semantic similarity is not proof an answer is supported by the retrieved source material.
C:Supervised production quality requires reliable prediction-to-ground-truth linkage, not only unlabeled model outputs. This principle describes ground truth association for quality monitoring, which is not the cause of the observed generative ai factual grounding evaluation failure. The controlling explanation is: Fluent text or semantic similarity is not proof an answer is supported by the retrieved source material.
D:Excess prompt context can increase processing cost and latency without improving response accuracy. This principle describes llm latency and cost token regression, which is not the cause of the observed generative ai factual grounding evaluation failure. The controlling explanation is: Fluent text or semantic similarity is not proof an answer is supported by the retrieved source material.
Question 620
During a regulated MLOps pipeline deployment, the following condition occurs: A fraud classifier is retrained every night despite true chargeback outcomes arriving weeks later, producing highly incomplete recent labels. The workflow must be repeatable and protect the last approved production model. What acceptance result verifies the intended behavior after remediation?
- No-op periods create no unnecessary models and validated drift events produce candidates requiring approval.
- Training manifests exclude unlabeled recent transactions and measured metrics use verified ground truth.
- An audit can resolve model version to exact source data and processing code without guessing by timestamps.
- A previous model can be rebuilt from its exact documented training dataset and code revision.
Explanation
Correct Answer
B:Training manifests exclude unlabeled recent transactions and measured metrics use verified ground truth. This is specific acceptance evidence for scheduled retraining on delayed ground truth. Unobserved outcomes cannot serve as reliable supervised labels regardless of the desired retraining schedule. It confirms the controlled correction: Train only on adequately matured labeled cohorts and monitor recent feature drift separately until outcomes arrive.
Incorrect Answers
A:No-op periods create no unnecessary models and validated drift events produce candidates requiring approval. This result could demonstrate success for data drift triggered retraining policy but not for the stated scheduled retraining on delayed ground truth issue. The direct test is: Training manifests exclude unlabeled recent transactions and measured metrics use verified ground truth.
C:An audit can resolve model version to exact source data and processing code without guessing by timestamps. This result could demonstrate success for model feature lineage for audit but not for the stated scheduled retraining on delayed ground truth issue. The direct test is: Training manifests exclude unlabeled recent transactions and measured metrics use verified ground truth.
D:A previous model can be rebuilt from its exact documented training dataset and code revision. This result could demonstrate success for data version reproducibility in ct jobs but not for the stated scheduled retraining on delayed ground truth issue. The direct test is: Training manifests exclude unlabeled recent transactions and measured metrics use verified ground truth.