Create a custom practice set
Pick category, difficulty, number of questions, and time limit. Start instantly with your own quiz.
Generate QuizGK Questions
Filter by category, type, and difficulty. Reading is open for everyone.
The computing paradigm that uses quantum mechanical phenomena for solving optimization problems is called ________ Annealing.
Answer: Quantum
Quantum annealing uses quantum tunneling to find global minima in optimization landscapes. Suitable for logistics, finance, machine learning. Different from gate-based quantum computing.
The technique that replaces sensitive data with non-sensitive equivalents while preserving format is called ________.
Answer: Tokenization
Tokenization replaces sensitive values with format-preserving tokens, reversible via secure vault. Used in payments, healthcare for PCI-DSS compliance.
The Right to Erasure allows individuals to request deletion of their personal data under DPDP Act.
- True
- False
Answer: True
Right to Erasure (Right to be Forgotten) allows individuals to request deletion when data no longer necessary or consent withdrawn. Critical for privacy compliance.
The technique that reconstructs sensitive attributes from model predictions is called Model ________ Attack.
Answer: Inversion
Model inversion exploits model outputs to infer sensitive training data attributes. Critical privacy risk for ML systems handling personal data.
Data poisoning attacks manipulate training data to compromise model behavior during inference.
- True
- False
Answer: True
Data poisoning injects malicious samples into training data to cause targeted misclassifications or backdoors. Mitigation: data validation, robust training, anomaly detection.
Which technique protects ML models from adversarial attacks by training with perturbed inputs?
- Adversarial Training
- Defensive Distillation
- Both A and B
- Neither
Answer: Both A and B
Adversarial training includes adversarial examples in training data; defensive distillation smooths model outputs. Both improve robustness against adversarial attacks.
The pattern of gradually shifting traffic between application versions for safe deployments is called ________ Release.
Answer: Canary
Canary Release deploys new version to small user subset, monitors metrics, gradually expands if stable. Reduces risk vs big-bang releases.
The Indian mission that successfully demonstrated in-orbit docking technology in 2024 is called ________.
Answer: SPADEX
SPADEX (Space Docking Experiment) successfully demonstrated autonomous rendezvous and docking of two spacecraft in Earth orbit.
The Hyderabad-based initiative that is India's largest startup incubator is called ________.
Answer: T-Hub
T-Hub (Telangana) is India's largest innovation campus, connecting startups with corporates, investors, mentors. Supports deep tech, healthtech, enterprise solutions.
Which DeFi protocol enables automated market making without order books?
- Uniswap
- Aave
- Compound
- MakerDAO
Answer: Uniswap
Uniswap uses automated market maker (AMM) model with liquidity pools instead of order books. Enables permissionless token trading with constant product formula.
Which technology enables real-time monitoring of renewable energy generation for grid balancing?
- SCADA
- IoT Sensors
- AI Forecasting
- All of these
Answer: All of these
Renewable integration combines SCADA for control, IoT sensors for real-time data, AI forecasting for generation prediction. Critical for grid stability with high renewable penetration.
Which ISRO mission is designed to study Venus's atmosphere and surface?
- Shukrayaan-1
- Mangalyaan-2
- Aditya-L2
- Gaganyaan-2
Answer: Shukrayaan-1
Shukrayaan-1 is India's planned Venus orbiter mission (target launch ~2028) to study the planet's thick atmosphere, surface processes, and solar wind interactions.
The frequency band used for high-bandwidth satellite internet with higher rain attenuation is ________-band.
Answer: Ka
Ka-band (26.5-40 GHz) offers high bandwidth for satellite broadband but susceptible to rain attenuation. Requires adaptive coding and ground station diversity.
Which computing paradigm mimics brain architecture for energy-efficient AI processing?
- Quantum Computing
- Neuromorphic Computing
- Optical Computing
- Analog Computing
Answer: Neuromorphic Computing
Neuromorphic chips (Intel Loihi, IBM TrueNorth) use spiking neural networks and event-driven processing for orders-of-magnitude better energy efficiency than GPUs for AI tasks.
Which quantum algorithm threatens RSA encryption by factoring large integers efficiently?
- Grover's Algorithm
- Shor's Algorithm
- Quantum Fourier Transform
- Variational Quantum Eigensolver
Answer: Shor's Algorithm
Shor's Algorithm factors integers in polynomial time on quantum computers, breaking RSA which relies on factoring difficulty. Requires fault-tolerant quantum computers not yet available.
The authority that adjudicates disputes under the DPDP Act 2023 is called Data Protection ________ of India.
Answer: Board
Data Protection Board of India is independent regulatory body established under DPDP Act 2023 to adjudicate disputes, impose penalties, enforce compliance.
The DPDP Act 2023 applies extraterritorially to processing related to Indian individuals.
- True
- False
Answer: True
DPDP Act governs processing of digital personal data: (1) within India, and (2) outside India if offering goods/services or profiling individuals in India.
The technique that adds calibrated noise to model outputs to prevent membership inference is called ________ Privacy.
Answer: Differential
Differential Privacy mathematically guarantees that model outputs don't reveal whether specific individuals were in training data. Critical for privacy-preserving ML deployment.
Model inversion attacks reconstruct training data from model outputs, posing privacy risks.
- True
- False
Answer: True
Model inversion exploits model predictions to infer sensitive training data attributes. Mitigation: differential privacy, output perturbation, access controls. Critical for privacy-preserving ML.
Which attack manipulates ML model inputs to cause misclassification while appearing normal to humans?
- Data Poisoning
- Adversarial Examples
- Model Inversion
- Membership Inference
Answer: Adversarial Examples
Adversarial examples add imperceptible perturbations to inputs that cause ML models to misclassify. Critical vulnerability for computer vision, NLP systems requiring robust defenses.