Description:

Explainability approaches for deep learning have gained significant importance in recent years due to the widespread adoption of deep neural networks in various applications. Despite their remarkable performance, deep learning models are often perceived as black boxes, lacking transparency and interpretability. This lack of interpretability raises concerns regarding the reliability and trustworthiness of deep learning predictions, especially in critical domains such as healthcare, finance, and autonomous systems. To address these issues, a plethora of explainability methods have been proposed, aiming to provide insights into the decision-making process of deep learning models. We will present an overview of explainability approaches for deep learning. Additionally, we explore the trade-offs between interpretability and model performance and discuss the challenges and future directions in the field of deep learning explainability. Understanding these approaches is essential to enhance the transparency and trustworthiness of deep learning models, fostering their application in real-world scenarios while ensuring human-centric AI development.

Outline-

• ML models are not 100% accurate
• Also, they are very much like black-box models
• At times, their decisions need to be reviewed by human agent
• But if they can explain their decisions then the human in the loop (BA) will get more confidence about model’s behavior
• This will also help us train better models by giving feedback to the ML model designer and prevent adversarial attacks

Prerequisites:

Python, AI, Deep Learning

Speaker Info:

Speaker 1: Sai Babu Udayagiri

Title: Data Science Lead - Applied AI Research

Company/Organization: Societe Generale

Biography:

Sai Babu Udayagiri is a distinguished Data Science Lead with extensive experience in building Machine Learning (ML) systems and driving applied AI research in the industry. With a successful career spanning over 5 years, Sai has established themselves as a leading expert in the application of AI across various domains.

As the Data Science Lead at Societe Generale, Sai plays a crucial role in spearheading applied AI research projects. Leveraging their expertise in ML techniques and data-driven insights, Sai contributes significantly to developing cutting-edge solutions for complex business challenges. Their work has had a transformative impact on optimizing processes, enhancing decision-making, and driving innovation within the organization.

With a keen interest in the intersection of AI and industry-specific domains, Sai has successfully led projects in diverse areas such as predictive analytics, computer vision, and natural language processing (NLP). Their ability to harness the power of data and deploy advanced ML models has earned them a reputation for delivering impactful solutions in real-world scenarios.

Sai is a sought-after speaker and thought leader in the field of applied AI research. They have delivered engaging talks and workshops at renowned conferences and industry events, sharing valuable insights into the practical applications of AI and ML in various industries.

In addition to their industry contributions, Sai is committed to fostering collaboration between academia and industry. Their dedication to advancing the responsible and ethical use of AI ensures that AI technologies are developed and deployed with a human-centric approach.

Join Sai Babu Udayagiri in an enlightening talk as they explore the exciting world of applied AI research. Gain valuable insights from their wealth of industry experience and learn how AI-driven solutions are transforming businesses and industries. Be inspired by their journey of leveraging ML techniques to drive innovation and solve real-world challenges in the dynamic landscape of applied AI.

Speaker 2: Yash Bhatnagar

Title: Associate Data Scientist - Applied AI Research

Company/Organization: Societe Generale

Biography: Yash is an Associate Data Scientist from Societe generale Global Solutions Centre (a French multinational investment Bank) with 2+ years of experience in the field of AI and ML. His team builds Machine Learning driven enterprise solutions for the various business lines of the Bank. His responsibilities include leading project teams consisting of software engineers, designers and data scientists, presenting achievements to senior stakeholders and navigating feature development of products. He is based out of Bengaluru, and is a B.Tech graduate from IIT Delhi.

Speaker 3: Kaushal Kishore

Title: Software Engineer

Company/Organization: Societe Generale

Biography:

Kaushal Kishore is a software engineer working in Deep learning and AI domain at Societe Generale GSC. He is a graduate from IIT Bombay and has immense enthusiasm to explore the cutting edge technological advancements and contribute to it.

Speaker Links:

Sai Babu Udayagiri- https://in.linkedin.com/in/dr-sai-babu-ph-d-38164a3a

Yash Bhatnagar- https://in.linkedin.com/in/yash-bhtngr

Kaushal Kishore- github id : kaushore linkedin id: kaushore