Title: Principles of Learning from Multiple Tasks Abstract:  Deep networks are increasingly trained on data from multiple tasks with the goal of sharing synergistic information across related tasks. A language model,…

Speaker: Christopher A. Baldassano Title: Remembering events using schematic knowledge Abstract: Our everyday experiences consist of familiar sequences of events in familiar contexts, and we use our knowledge of the past to understand…

Title: How the human brain thinks about itself Abstract: The human brain has a remarkable ability to monitor and evaluate its own mental states, known as metacognition. Metacognition is crucial to success, enabling us to recognise gaps in our knowledge and collaborate effectively. Problems with metacognition are linked to maladaptive behaviours, such as endorsing false…

Title: Quantifying individuality in neural circuit representations   Abstract: Signatures of neural computation are thought to be reflected in the coordinated activity of large neural populations. Neuroscience is now flush with measurements of these activity patterns in humans, animal subjects, and large-scale artificial network models. In this talk, I will address an extensively studied, yet unresolved, question: How should we quantify the extent to which…

LLM Benchmarks Zoom link: https://columbiauniversity.zoom.us/j/97176853843?pwd=VLZdh6yqHBcOQhdf816lkN5ByIpIsF.1

Title: Reimagining the biology of memory   Abstract: Over the last half century, there has been a remarkable convergence on the idea that memories are stored at synapses. I will argue that this is only part of the story. A more complete story compels us to recognize the radical ubiquity of memory in living systems,…

Bio: Itzel is a Ph.D. student at Rice University working with Prof. Xaq Pitkow. She studies perception and control mechanisms that give biological organisms an advantage over machines. She believes understanding how the brain works using mathematical principles is essential to build the next generation of AI systems which are more robust, more general-purpose, less…

Title: Uncovering latent low-dimensional structure in network connectivity Abstract: Network connectivity constrains the patterns of neural activity in the brain. These constraints are often observed as low-dimensional manifolds in neural activity space. Continuous Attractor Networks (CANs) are a prime example of this type of network phenomenon. Interestingly, there are examples of CANs where the structure or topology…

Zoom link: https://columbiauniversity.zoom.us/j/97176853843?pwd=VLZdh6yqHBcOQhdf816lkN5ByIpIsF.1

Title: The structure of individuality in micro-behavioral features of task performance Abstract: Individuality is an intrinsic and essential aspect of mammalian behavior that emerges even in genetically identical organisms experiencing the same environmental conditions. In the International Brain Laboratory (IBL),  mice were trained on a visual decision-making task with the explicit goal of establishing a…

This event is for all ARNI trainees and graduate students who work on an ARNI related project. The goal of this symposium is to foster networking and career development. Location: Zuckerman Institute L3-079 Time: 12pm to 5pm Registration form: https://forms.gle/4e2AHqP54X8VvkKS8 Program 12pm: Lunch Catering by FUMO 1pm: ARNI Postdoc Presentations Haozhe Shan Mehdi Azabou 2pm:…

The working group will discuss: https://www.nature.com/articles/s41593-020-0671-1 Join via Google Meets: meet.google.com/nnq-csiy-yah