Institute for Pure & Applied Mathematics (IPAM) Courses

Informational and Topological Signatures of Individuality in Network Neuroscience

Exploring brain function through topological data analysis, demonstrating improved individual and age discrimination using algebraic-topological features extracted from resting fMRI data compared to traditional methods.

• YouTube
• 45 minutes
• Self-Paced
• Free Video

Using Dynamics to Constrain Connectivity of Neuronal Population in Respiratory Brainstem

Explores models of neuronal connectivity in respiratory brainstem, focusing on synchronized bursting activity and network architecture insights using multi-state bootstrap percolation.

• YouTube
• 53 minutes
• Self-Paced
• Free Video

Analysis of Cortical Connectomes

Exploring methods for analyzing complex cortical connectomes, including Bayesian circuit modeling and clustering approaches, to understand neuronal circuits in mammalian cerebral cortex.

• YouTube
• 51 minutes
• Self-Paced
• Free Video

Connectome-Based Modeling in the Fly

Explore connectome-based modeling in flies, revolutionizing neural circuit understanding and its link to behavior, presented by Larry Abbott at UCLA's IPAM workshop.

• YouTube
• 49 minutes
• Self-Paced
• Free Video

On the Hardness of Learning Under Symmetries

Explore the challenges of machine learning with symmetries in this talk by MIT's Thien Le, examining computational and statistical gaps in learning and optimization.

• YouTube
• 47 minutes
• Self-Paced
• Free Video

Understanding and Overcoming the Statistical Limitations of Decision Trees

Explores decision trees' statistical limitations, proposes FIGS algorithm to improve performance, and demonstrates its effectiveness in clinical decision-making and other real-world applications.

• YouTube
• 45 minutes
• Self-Paced
• Free Video

Complexity of Adversarially Robust Proper Learning of Halfspaces with Agnostic Noise

Explore computational complexity of robust halfspace learning with agnostic noise, focusing on L_p perturbations. Discover efficient algorithms and hardness results for distribution-independent PAC models.

• YouTube
• 49 minutes
• Self-Paced
• Free Video

Smoothed Analysis for Learning Concepts with Low Intrinsic Dimension

Explore smoothed analysis in machine learning, focusing on low-dimensional concepts and their applications. Gain insights into new algorithms for agnostic learning and margin-based approaches.

• YouTube
• 44 minutes
• Self-Paced
• Free Video

Algorithmic Fairness, Loss Minimization and Outcome Indistinguishability

Explore algorithmic fairness, loss minimization, and outcome indistinguishability in machine learning. Discover how these concepts address fairness issues and enable flexible predictor optimization across various constraints.

• YouTube
• 51 minutes
• Self-Paced
• Free Video

Learning from Aggregate Responses

Explore aggregate learning techniques for privacy-protected data, including loss construction, bagging schemes, and iterative boosting algorithms to improve model accuracy while maintaining data privacy.

• YouTube
• 45 minutes
• Self-Paced
• Free Video

New Local Differentially Private Protocols for Frequency and Mean Estimation

Explore new protocols for frequency and mean estimation in local differential privacy, addressing privacy concerns in distributed applications and large-scale data analysis.

• YouTube
• 49 minutes
• Self-Paced
• Free Video

Learning-Augmented Online Optimization

Explore learning-augmented online optimization with Google's Ravi Kumar, covering computational and statistical gaps in learning and optimization techniques.

• YouTube
• 49 minutes
• Self-Paced
• Free Video

Learning from Dynamics in Linear Dynamical Systems

Explore efficient algorithms for learning linear dynamical systems, bridging gaps in understanding and revealing connections between machine learning tools and non-stationary settings.

• YouTube
• 48 minutes
• Self-Paced
• Free Video

A Perceptron Trio - Computational vs Statistical Gaps in Learning and Optimization

Explore perceptron algorithms in machine learning, focusing on computational and statistical aspects in optimization and learning theory.

• YouTube
• 51 minutes
• Self-Paced
• Free Video

Is It Easier to Count Communities Than Find Them?

Explore computational challenges in detecting community structures in random graphs, examining statistical and computational phase transitions, and the difficulty of inferring community properties.

• YouTube
• 47 minutes
• Self-Paced
• Free Video