What’s in today’s article?

• Why in News?
• What is India's AI Mission?
• 7 Key Features of India AI Mission
• Significance of India AI Mission
• Changes Introduced by the MeitY under its AI Mission

Why in News?

• Under its ambitious AI Mission, the Ministry of Electronics and IT (MeitY) has relaxed some provisions in its norms to procure computing capacity for artificial intelligence (AI) solutions, addressing concerns from smaller companies.

What is India’s AI Mission?

• The PM of India at the Global Partnership on Artificial Intelligence (GPAI) Summit 2023 (New Delhi) announced that India will launch an artificial intelligence (AI) mission.
• Under the India AI Mission [to be implemented by the ‘IndiaAI’ Independent Business Division (IBD) under Digital India Corporation (DIC)], the MeitY will look -
  • To establish a computing capacity of more than 10,000 graphics processing units (GPUs), which is an electronic circuit that processes images and accelerates the rendering of 3D computer graphics, and
  • To help develop foundational models trained on datasets covering major Indian languages for priority sectors like healthcare, agriculture, and governance.
• AI Curation Units (ACUs) will also be developed in 50-line ministries and AI marketplace will be designed to offer AI as a service and pre-trained models to those working on AI applications.
• The implementation of this AI compute infrastructure will be done through a public-private partnership model with 50% viability gap funding.
• Of the total outlay (of Rs 10,372 crore), Rs 4,564 crore has been earmarked for building computing infrastructure.

7 Key Features of India AI Mission:

• India AI Compute Capacity: This pillar will build a high-end scalable AI computing ecosystem to cater to the increasing demands from India’s rapidly expanding AI start-ups and research ecosystem.
• India AI Innovation Centre: The Centre will undertake the development and deployment of indigenous Large Multimodal Models (LMMs) and domain-specific foundational models in critical sectors.
• India AI Datasets Platform: This will streamline access to quality non-personal datasets for AI Innovation.
• India AI Application Development Initiative: This will promote AI applications in critical sectors for the problem statements sourced from Central Ministries, State Departments, etc.
• India AI FutureSkills: It is conceptualised to mitigate barriers to entry into AI programmes and will increase AI courses in undergraduate, Masters level, and Ph.D. programmes.
• India AI Startup Financing: This pillar is conceptualised to support and accelerate deep-tech AI startups and provide them with streamlined access to funding to enable futuristic AI projects.
• Safe & Trusted AI: This pillar will enable the implementation of responsible AI projects including the development of indigenous tools and frameworks.

Significance of India AI Mission:

• The India AI Mission will further the vision of Making AI in India and Making AI Work for India.
• It seeks to showcase the international community the positive applications of the revolutionary technology for societal benefit, thereby elevating India's global competitiveness.
• It will establish a comprehensive ecosystem catalysing AI innovation through strategic programmes and partnerships across the public and private sectors.
• It will drive forward creativity and enhance internal capabilities, safeguarding India's technological autonomy.
• Additionally, it will generate employment opportunities that require advanced skills, tapping into the country's demographic advantage.

Changes Introduced by the MeitY under its AI Mission:

• Key changes include:
  • Lowered annual turnover requirements:
    • For primary bidders, it has been reduced from Rs 100 crore to Rs 50 crore, and
    • For non-primary members, it has been halved to Rs 25 crore.
• This adjustment aims to include more start-ups in the bidding process.
  • Reduced computing capacity requirements: For instance, the requirement for FP16 performance has been cut from 300 TFLOPS to 150 TFLOPS, and AI compute memory has been reduced from 40 GB to 24 GB.
  • Experience criteria: Companies must now demonstrate prior experience in providing AI services, including the number of clients served and minimum billing of Rs 10 lakh over the past three financial years.
  • Local sourcing mandate: Bidders are required to source components for cloud services from Class I or Class II local suppliers, promoting the 'Make in India' initiative.
  • Data sovereignty: All AI services must be delivered from data centers within India, ensuring that user data remains within the country's borders.
  • Public-Private Partnership (PPP) model: The implementation will follow a PPP model, with 50% viability gap funding. A significant portion of the budget (Rs 4,564 crore) is allocated for building computing infrastructure.
• Significance of these changes: These changes are expected to facilitate greater participation from smaller firms, enhancing the AI landscape in India.

Q.1. What is data sovereignty?

Data sovereignty is the idea that data is subject to the laws of the country or region where it's generated or stored. It's important because it protects data from unauthorized access and breaches, and ensures businesses have access to their data in case of a disaster.

Q.2. Which law governs data sovereignty in India?

India's data sovereignty laws are governed by the Digital Personal Data Protection Act (DPDPA), which was passed by the Indian Parliament in 2023.

Source: To accommodate start-ups, MeitY relaxes AI compute procurement norms

What’s in today’s article?

• Why in News?
• Space Debris
• PSLV- C37 upper stage re-enters Earth orbit

Why in News?

ISRO announced the successful re-entry of the Polar Satellite Launch Vehicle-C37 (PSLV-C37) upper stage into Earth's atmosphere on October 6. This achievement aligns with global efforts to reduce space debris, supporting ISRO's goal of achieving debris-free space missions by 2030.

Space Debris

• About
  • Space debris refers to defunct, human-made objects in Earth's orbit, such as non-functional satellites, spent rocket stages, and fragments from satellite collisions.
  • These debris pose a growing threat to operational spacecraft, satellites, and the International Space Station (ISS).
• Challenges
  • Collisions: Even tiny debris can cause severe damage to satellites and space stations due to their high velocities.
  • Chain Reaction (Kessler Syndrome): Increasing debris can lead to more collisions, creating even more debris and increasing the risk of cascading damage.
  • Cost of Mitigation: Tracking and removing debris require advanced technology and substantial financial investment.
• Increasing space debris
  • With the rise in the number of satellites in orbit around the earth, space debris has become a pressing issue.
  • According to ISRO’s Space Situational Assessment report 2022, the world placed 2,533 objects in space in 179 launches in 2022.
  • The number of space objects greater than 10 cm in size in LEO is expected to be about 60,000 by 2030.
• Legal provisions
  • Currently, there are no international space laws pertaining to LEO debris.
  • However, most space-exploring nations abide by the Space Debris Mitigation Guidelines 2002 specified by the IADC. This was endorsed by the U.N. in 2007.
  • The guidelines outline methods to limit accidental collisions in orbit, break-ups during operations, intentional destruction, and post-mission break-ups.
• International Institutions
  • Inter-Agency Space Debris Coordination Committee (IADC): A global forum that coordinates efforts to mitigate space debris.
  • United Nations Committee on the Peaceful Uses of Outer Space (COPUOS): Sets space debris mitigation guidelines.
  • International Telecommunication Union (ITU): Regulates satellite orbital slots to prevent overcrowding.
• Steps taken by India
  • Debris Free Space Mission (DFSM): ISRO is committed to achieving a debris-free space environment by 2030 through passivation, active de-orbiting, and controlled re-entry of spent rocket stages.
  • The implementation of this DFSM initiative will start by the beginning of 2025.
  • This includes selecting clean orbits, budgeting fuel for post-mission disposal, and precisely controlling re-entry trajectories.
  • ISRO’s Debris Mitigation Strategy: India’s space agency, ISRO, follows international guidelines, including the IADC's recommendation to limit post-mission orbital life to 25 years.
  • IS4OM (ISRO System for Safe and Sustainable Space Operations Management): Monitors space debris and orbital decay, ensuring compliance with mitigation guidelines.

PSLV- C37 upper stage re-enters Earth orbit

• Re-entry of PSLV- C37 upper stage
  • On October 6, 2024, the upper stage of ISRO's PSLV-C37 mission, launched in February 2017, re-entered Earth's atmosphere.
• The PSLV-C37 carried 104 satellites, including Cartosat-2D as the primary payload.
  • After the mission, the upper stage (PS4) remained in orbit at approximately 470x494 km.
  • Over time, its orbit decayed due to atmospheric drag and was closely monitored by ISRO and US Space Command (USSPACECOM).
• Re-entry followed international debris mitigation guidelines
  • This re-entry followed international debris mitigation guidelines, specifically the IADC recommendation to limit the post-mission orbital life to 25 years.
  • ISRO's passivation sequence successfully lowered PS4’s orbit, ensuring re-entry within eight years.
  • ISRO is now working to further reduce the orbital lifetime of rocket stages to five years through active de-orbiting, with future missions focusing on controlled re-entry.
• ISRO also aims to achieve a Debris Free Space Mission (DFSM) by 2030.

Q.1. What is space debris and why is it a concern?

Space debris consists of defunct objects in orbit, posing risks of collisions and cascading damage, impacting space operations and safety.

Q.2. How is ISRO addressing the space debris issue?

ISRO is implementing strategies like active de-orbiting and controlled re-entry to reduce space debris, with a goal of achieving debris-free space missions by 2030.

Source: Upper stage of historic PSLV-37 mission re-enters Earth’s atmosphere eight years after launch: ISRO | ISRO | Indian Express

What’s in today’s article?

• Why in News?
• What is Machine learning (ML)?
• What is Deep Learning (DL)?
• What is Artificial Neural Network (ANN)?
• Works of Noble Prize winners

Why in News?

The 2024 Nobel Prize in Physics was awarded to John Hopfield and Geoffrey Hinton for their foundational contributions to AI, particularly in machine learning and artificial neural networks.

Their ground-breaking research in the 1980s laid the foundation for the AI revolution unfolding today.

What is Machine learning (ML)?

• About
  • ML is a subset of artificial intelligence (AI) that enables computers to learn from and make decisions based on data without being explicitly programmed for each task.
  • In machine learning, algorithms identify patterns in large datasets and use these patterns to make predictions or perform specific tasks.
  • The key idea is that systems improve their performance over time through experience, by training on data.
• Applications of Machine Learning:
  • Image and speech recognition
  • Recommendation systems (like those used by streaming services)
  • Fraud detection
  • Healthcare diagnostics
  • Autonomous vehicles

What is Deep Learning (DL)?

• About
  • Deep Learning is a specialized subset of machine learning that focuses on using artificial neural networks with multiple layers (hence "deep").
  • It mimics the structure and function of the human brain to recognize complex patterns in large datasets, such as images, text, or sound.
  • Deep learning has been pivotal in advancing AI technologies, particularly in areas like image recognition, natural language processing, and self-driving cars.
• Key Applications of Deep Learning:
  • Image and speech recognition (e.g., face detection, virtual assistants)
  • Autonomous vehicles (e.g., self-driving cars)
  • Natural language processing (e.g., language translation)
  • Medical diagnostics (e.g., cancer detection in medical imaging)
• ML Vs. DL
  • While machine learning involves training algorithms with structured data and often requires human input for feature extraction, deep learning automates feature discovery using multi-layered neural networks, making it more powerful for complex tasks, especially when large datasets are available.

What is Artificial Neural Network (ANN)?

• About
  • ANN is a mathematical model that uses a network of interconnected nodes to mimic the human brain's neurons and process data. 
  • ANNs are a type of machine learning (ML) and deep learning that can learn from mistakes and improve over time. 
  • They are used in artificial intelligence (AI) to solve complex problems, such as recognizing faces or summarizing documents. 
• Key features of ANNs
  • Structure
    • ANNs are made up of layers of nodes, each containing an activation function. The nodes are interconnected, with each node in a layer connected to many nodes in the previous and next layers.
  • Learning
    • ANNs are adaptive and learn from their mistakes using a backpropagation algorithm.
    • They modify themselves as they learn, with inputs that contribute to the right answers weighted higher.
  • Output
    • The output of the ANN is produced by the final layer of nodes. The output is usually a numerical prediction about the information the ANN received.
• Applications of Artificial Neural Networks:
  • Image and video recognition (e.g., facial recognition systems)
  • Speech recognition (e.g., virtual assistants like Siri and Alexa)
  • Natural language processing (e.g., language translation)
  • Medical diagnostics (e.g., detecting diseases from medical images)
  • Autonomous vehicles (e.g., self-driving car navigation)
• In essence, artificial neural networks mimic the brain’s ability to learn from experience, adapt, and recognize complex patterns, making them foundational to modern AI and machine learning systems.

Works of Noble Prize winners

• Hopfield's contribution - Mimicking the Brain with Neural Networks
  • Hopfield's major breakthrough was creating artificial neural networks that mimic human brain functions like remembering and learning.
  • Hopfield's network processes information using the entire structure rather than individual bits, unlike traditional computing.
  • It captures patterns holistically, such as an image or song, and recalls or regenerates them even from incomplete inputs.
  • This breakthrough advanced pattern recognition in computers, paving the way for technologies like facial recognition and image enhancement.
  • His research was inspired by earlier discoveries in neuroscience, notably Donald Hebb's work on learning and synapses in 1949.
• Hinton’s Contribution - Deep Learning and Advanced Neural Networks
  • Hinton advanced Hopfield’s work by developing deep neural networks capable of complex tasks like voice and image recognition.
  • His method of backpropagation enabled these networks to learn and improve over time through training with large datasets.
• Backpropagation, short for "backward propagation of errors," is an algorithm for supervised learning of artificial neural networks using gradient descent. 
  • His contributions led to major advancements in AI technologies, including modern applications such as speech recognition, self-driving cars, and virtual assistants.
  • Hinton's deep learning networks made a significant impact at the 2012 ImageNet Visual Recognition Challenge, where his team's algorithm dramatically improved image recognition technology.
  • His work demonstrated the vast potential of AI in various fields, including astronomy, where machine learning helps researchers analyze vast amounts of data.
• Conclusion
  • Both Hopfield and Hinton have made pioneering contributions to the development of AI, with Hopfield bridging neuroscience, physics, and biology, and Hinton revolutionizing computer science.
  • Their work has shaped modern AI technologies, making them deserving recipients of the Nobel Prize in Physics.

Q.1. What are the contributions of John Hopfield in AI?

John Hopfield developed artificial neural networks that mimic the brain, enabling computers to recognize and learn patterns, a breakthrough in AI.

Q.2. How did Geoffrey Hinton advance deep learning?

Geoffrey Hinton introduced deep neural networks and backpropagation, revolutionizing AI tasks like speech and image recognition through continuous learning.

Source: Making machines learn | Nature | AWS Amazon | IBM