BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CMSA - ECPv6.15.18//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:CMSA
X-ORIGINAL-URL:https://cmsa.fas.harvard.edu
X-WR-CALDESC:Events for CMSA
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20250309T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20251102T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20260308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20261101T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20270314T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20271107T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20260427T090000
DTEND;TZID=America/New_York:20260501T170000
DTSTAMP:20260410T174150
CREATED:20250724T152524Z
LAST-MODIFIED:20260410T204101Z
UID:10003757-1777280400-1777654800@cmsa.fas.harvard.edu
SUMMARY:Mathematics and Biology II: Mathematics and Science of Behavior
DESCRIPTION:Mathematics and Biology II: Mathematics and Science of Behavior \nDates: April 27 – May 1\, 2026 \nLocation: Harvard CMSA\, Room G10\, 20 Garden Street\, Cambridge MA \n\n\nThis meeting will explore the emerging mathematics and science of embodied cognition—the idea that behavior arises not solely from the brain but through the dynamic interaction of brain\, body\, and environment. Understanding how animals sense\, move\, decide\, and coordinate\, from individual sensorimotor loops to collective dynamics\, demands mathematical frameworks that integrate geometry\, dynamics\, stochastic processes\, control theory\, and multiscale physics. The meeting will bring together experimentalists studying behavior across species with theorists and engineers building mathematical models and bio-inspired machines\, to identify shared principles of adaptive behavior. \n\n\nCo-organizers: L. Mahadevan (Harvard)\, Francesco Mori (Harvard CMSA)\, Venkatesh Murthy (Harvard) \nRegister to attend in-person \n  \nSpeakers \n\nPulkit Agrawal\, MIT\nKristin Branson\, HHMI\nAntonio C. Costa\, Sorbonne University/Paris Brain Institute\nNoah Cowan\, Johns Hopkins University\nRobert Datta\, Harvard Medical School\nBen de Bivort\, Harvard University\nOfer Feinerman\, Weizmann Institute of Science\nDeborah Gordon\, Stanford University\nAlbert Kao\, UMass Boston\nAnn Kennedy\, Scripps Research Institute\nHungtang Ko\, Tufts University\nGeorge Lauder\, Harvard University\nBence Ölveczky\, Harvard University\nKirstin Petersen\, Cornell University\nPavan Ramdya\, EPFL\nElizabeth Tibbetts\, University of Michigan\nRobert Wood\, Harvard University\n\n  \n\nSchedule  \nMonday\, Apr. 27\, 2026 \n9:00–9:30 am: Breakfast \n9:30–10:15 am: Deborah Gordon\, Stanford University\nTitle: The dynamics of collective behavior in changing environments\nAbstract: Collective behavior operates without central control\, using interactions among participants adjust to changing conditions. There is enormous diversity in the dynamics of collective behavior\, including in the rate of response to conditions\, in feedback regimes that set whether interactions stimulate or inhibit activity\, and the extent of centralization or modularity of information flow. An ecological perspective suggests how this diversity of collective behavior reflects the dynamics of the environment\, including its stability\, the ratio of resources spent to resources gained\, and the distribution of resources in time and space.\nAs examples\, I will discuss field studies and modelling of the regulation of foraging behavior in two species of ants\, Harvester ant colonies in the desert regulate foraging to manage high costs\, in water loss\, to obtain scattered and stable resources. They use a centralized system\, with the default to remain inactive unless stimulated\, that is slow to adjust foraging activity. In contrast\, the turtle ant colonies form trail networks in the canopy of the tropical forest\, in unstable conditions where activity costs are low\, to find and collect ephemeral and patchy resources. They use a highly modular system\, with the default to sustain activity unless inhibited\, that can rapidly adjust trail networks to changing resources and conditions. \n10:15–10:30 am: Discussion \n10:30–11:00 am: Tea Break \n11:00–11:45 am: Hungtang Ko\, Tufts University\nTitle: Collective mechanical intelligence: how fluid environments mediate self-organization of swarms\nAbstract: Biological collectives across scales self-organize within fluid environments. The mechanical coupling between swarming agents and fluid fields provides opportunities for both passive self-assembly and active\, fluid-mediated communication. However\, while sporadic evidence of collective mechanical intelligence exists\, its underlying mechanisms remain elusive. In this talk\, I will focus on two key systems: fire ant rafts and fish schools. Using a combination of experiments and mathematical models\, I will show that fire ant rafts leverage passive interfacial forces for self-assembly and self-stabilization. Furthermore\, I will demonstrate how schools of giant danio utilize mechanically intelligent formations in 3D\, and discuss how swarm robotics may provide the key to future research in collective mechanical intelligence. \n11:45 am–12:00 pm: Discussion \n12:00–1:30 pm: Catered Lunch \n1:30–2:15 pm: Albert Kao\, UMass Boston \n2:15–2:30 pm: Discussion \n2:30–3:15 pm: Ann Kennedy\, Scripps Research Institute\nTitle: Neural mechanisms that gate the expression of motivated behaviors \n3:15–4:30 pm: Discussion \n4:30–5:30 pm: CMSA Colloquium: Ofer Feinerman\, Weizmann Institute of Science \n  \nTuesday\, April 28\, 2026 \n9:00–9:30 am: Breakfast \n9:30–10:15 am: Ben de Bivort\, Harvard University\nTitle: Bayesian Inference on biophysical models of connectomes\nAbstract: Recent progress in connectomics has opened new frontiers for understanding the underlying principles of neural circuits. By leveraging high-resolution maps of synaptic connections\, computational models can simulate neural dynamics with unprecedented detail. However\, it remains challenging to parsimoniously integrate circuit activity data with connectomic information to make biological in- sights. We propose a Bayesian framework as a principled method for bringing to bear existing data\, enabling uncertainty quantification for inferring parameters of interest\, as well as for predicted circuit outputs. To demonstrate this approach\, we implement a simple spiking neuron model using leaky- integrate-and-fire dynamics in the Drosophila olfactory circuit\, incorporating available firing rate data. We evaluate how models with varying levels of biological detail fit experimental data and examine how training on different subsets of data influences model predictions. \n10:15–10:30 am: Discussion \n10:30–11:00 am: Trainee talk: Yasuki Isoe \n11:00–11:30 am: Trainee talk: Trainee Talk: Siddharth Jayakumar \n11:30 am–12:00 pm: Discussion \n12:00–1:30 pm: Catered Lunch \n1:30–2:15 pm: Noah Cowan\, Johns Hopkins University\nTitle: Toward a Control Theory for Active Sensing\nAbstract: Active sensing is often defined as “movement for the purpose of sensing.” Here\, I take a different perspective—that active sensing in biological systems is not a distinct class of behaviors\, but rather a set of movement phenomena that arise from a control objective. Biological sensors adapt to persistent stimuli\, acting like high-pass filters that tend to block “DC.” Such “change-detecting” sensors can support efficient coding with a high dynamic range\, and in engineering\, bio-inspired event cameras are similar: they transmit information only when a pixel changes and\, as such\, are extremely fast and make efficient use of bandwidth for the right applications. However\, such “AC” sensors pose technical challenges for control. Specifically\, event-like biological sensors can cause a nonlinear system (1) to lose local linear observability\, and (2) to become impossible to stabilize about an equilibrium point (Biswas\, Sontag\, Cowan\, Eur J Control\, 2025). Active sensing behaviors must emerge for stable control\, even in the somewhat paradoxical setting where the task-level goal is to remain stationary. Here\, I will discuss my lab’s progress in analyzing how animals use active sensing behaviors to format sensory information\, enhancing observability and control. I will also present our efforts to formalize controller synthesis with event-like sensors. \n2:15–2:30 pm: Discussion \n2:30–3:15 pm: Robert Datta\, Harvard Medical School \n3:15–4:30 pm: Discussion \n  \nWednesday\, April 29\, 2026 \n9:00–9:30 am: Breakfast \n9:30–10:15 am: Kristin Branson\, HHMI\nTitle: How can generative AI help us understand animal behavior?\nAbstract: Understanding animal behavior at an algorithmic level — what animals attend to\, how they form internal world models\, goals\, and plans\, and how state maps to action — remains a central challenge in neuroethology. Large-scale behavioral experiments now produce trajectory datasets of extraordinary scale and complexity\, but existing approaches necessarily compress this complexity to just a few dimensions. We argue that generative AI offers a path toward rich\, query-able models of the data. We adapt transformer-based sequence modeling to multi-agent animal keypoint trajectories\, treating behavior forecasting as analogous to next-token prediction. Our agent-based network inputs biologically-motivated sensory representations and outputs the distribution of future pose velocities. We show that the model captures statistical properties of the behavioral distribution. We have built a Python library that encapsulates the complexity of transforms relating raw keypoints and model inputs and outputs to make these tools extensible by the NeuroAI community and accessible to theorists and experimentalists. Finally\, we argue that mechanistic interpretability methods allow us to query trained models through the natural framework of artificial neuroethology experiments. \n  \n10:15–10:30 am: Discussion \n10:30–11:00 am: Tea Break \n11:00–11:30 am: Trainee talk: Golnar Gharooni Fard \n11:30 am–12:00 pm: Trainee talk: Wenyi Zhang\, Harvard University\nTitle: Mechanisms of Setpoint Control in Drosophila Navigation System\nAbstract: Navigation provides a powerful system for studying how animals balance behavioral persistence with flexibility. During navigation\, fruit flies often default to fast straight walking (or “menotaxis”) in a barren environment\, maintaining a stable heading setpoint over a long period of time. Conversely\, when the local environment is enriched with sensory stimuli\, flies often explore the environment with more frequent heading changes\, either through directed steering driven by a sequence of updating setpoints\, or through undirected turning driven by temporarily lifting the setpoint control. Although this framework suggests a central role for the setpoint in guiding navigation\, the neural mechanisms for flexible setpoint control remain unclear.\nHere we identified h∆A\, a central complex cell type involved in setpoint control. In an aversive heat paradigm\, hΔA played an important role in the fly’s sensory-driven deviation from the menotactic goal direction. We characterized hΔA population activity and found that it carries two separable activity components: a bump-like signal that encodes a slowly varying travel-direction-related setpoint\, and a spatially uniform signal associated with turning. We further identified modulatory inputs to hΔA that shape h∆A activity. Together\, these results support a model in which short- and long-timescale setpoints compete for steering control\, and suggest a circuit mechanism by which flies balance directional persistence with flexible reorientation under changing sensory conditions. \n12:00–1:30 pm: Catered Lunch \n1:30–2:15 pm: Bence Ölveczky\, Harvard University\nTitle: Using neuro-biomechanical simulations to probe neural control of learned skills\nAbstract: The goal of my lab is to decipher the circuit logic by which the brain learns and controls motor skills. The standard mechanistic approach is to dissect the underlying circuits brain area-by-brain area\, inferring function by relating recordings and perturbations within each to behavior. This runs into fundamental problems in highly recurrent systems\, where activity in any one node is shaped by the dynamics of the whole\, a problem compounded by the fact that the circuits we probe control a complex biomechanical body and not measurable features of behavior. I will discuss these challenges and present results suggesting that neuro-biomechanical simulation\, leveraging advances in physics simulation and AI\, can offer a powerful alternative window into the neural circuits underlying learned skills. \n2:15–2:30 pm: Discussion \n2:30–3:15 pm: Pavan Ramdya\, EPFL\nTitle: Towards fly-inspired legged robots\nAbstract: I will discuss our efforts to build biologically-inspired legged robots using behavioral measurements\, neuromechanical simulations\, and anatomical studies of Drosophila melanogaster. \n3:15–4:30 pm: Discussion \n  \nThursday\, April 30\, 2026 \n9:00–9:30 am: Breakfast \n9:30–10:15 am: Pulkit Agrawal\, MIT \n10:15–10:30 am: Discussion \n10:30–11:00 am: Tea Break \n11:00–11:45 am: Antonio C. Costa\, Sorbonne University/Paris Brain Institute\nTitle: Unraveling the structure of behavioral variation: a dynamical approach to naturalistic data\nAbstract: Animal behavior varies widely\, both within the same individual over time and between individuals. While often overlooked\, this variation reflects hidden control variables and mechanisms that were shaped by evolution. For example\, variation in behavioral traits can help populations withstand environmental change\, while atypical motor patterns in neurological disorders may offer clues for personalized therapies. Comparing such complex behaviors is difficult. When dynamics are nonlinear and unfold over multiple timescales\, standard metrics based on summary statistics often miss meaningful differences. To address this\, we introduce a framework that encodes multiscale dynamics to compare behavior from data. By modeling nonlinear dynamics probabilistically (using transfer operators inferred from time-series data)\, we define a distance metric that captures behavioral differences across timescales. Tailored to finite\, noisy datasets\, our approach identifies principal axes of variation and enables rigorous clustering of individual trajectories. We demonstrate this framework in various biological systems\, including bacterial chemotaxis and larval zebrafish locomotion\, where the inferred axes of behavioral variation reflect underlying physiological variables and developmental histories. \n11:45 am–12:00 pm: Discussion \n12:00–1:30 pm: Catered Lunch \n1:30–2:15 pm: Elizabeth Tibbetts\, University of Michigan\nTitle: What paper wasps can teach us about the evolution of animal minds\nAbstract: Why do animals differ in their cognitive abilities? Some animals fail at apparently simple tasks\, while others have a remarkable capacity to collect\, retain\, and use information from the environment to guide their behavior. Although paper wasps brains are smaller than a grain of rice\, Tibbetts will show that wasps can perform seemingly complex behaviors like individual face recognition\, transitive inference\, social eavesdropping\, and concept learning. She will also describe experiments that take advantage of natural variation in behavior within and among wasp species to test how social interactions shape the development and evolution of cognitive abilities. \n2:15–2:30 pm: Discussion \n2:30–3:15 pm: Robert Wood\, Harvard University\nTitle: The Mechanical Side of Artificial Intelligence\nAbstract: Artificial Intelligence research typically focuses on perception\, learning\, and control methods to enable autonomous agents\, including robots\, to make and act on decisions in real-world scenarios. However\, even the most capable AI without a well-designed physical structure is of minimal use for canonical robotics tasks. Our research is focused on the design\, mechanics\, materials\, and manufacturing of novel robot platforms that make perception\, control\, or action easier or more robust for natural\, unstructured\, and often unpredictable environments. Key principles in this pursuit include bioinspired designs\, smart materials for novel sensors and actuators\, and the development of multi-scale\, multi-material manufacturing methods. This talk will illustrate this philosophy by highlighting the creation of three classes of robots with unique hardware challenges: bioinspired microrobots\, soft-bodied robots for manipulation\, and robots for interacting with delicate marine life. \n3:15–4:30 pm: Discussion \n  \nFriday\, May 1\, 2026 \n9:00–9:30 am: Breakfast \n9:30–10:15 am: George Lauder\, Harvard University\nTitle: Fish schooling behavior from kinematics to hydrodynamics to energetics\nAbstract: Do fish moving in a school reduce their energetic costs compared to swimming alone? If so\, how does collective motion reduce the energy needed to move? Only within the last two years have experimental studies directly demonstrated that fish swimming in a group have lower energy expenditure than solitary locomotion. Most studies of how fish move in a collective have focused on understanding the potential benefits of swimming in fixed relative positions. But recent experiments on fish schooling behavior have revealed that fish within the school are nearly constantly rearranging their relative positions. In this talk I will show how fish in a school can save energy even if they do not maintain fixed positions. Analyses of water flow patterns within fish schools have been used to resolve this “paradox” and show that fish movement within a school creates hydrodynamic shelters with zones of reduced flow velocity that nearby fish can take advantage of. \n10:15–10:30 am: Discussion \n10:30–11:00 am: Tea Break \n11:00–11:45 am: Kirstin Petersen\, Cornell University\nTitle: Harnessing Embodied Intelligence in Robot Collectives\nAbstract: In the Collective Embodied Intelligence Lab\, we study embodied intelligence as a complement to artificial intelligence in robot collectives. Our work spans scales and mechanisms\, from behaviors encoded in robot morphology to collective behaviors that emerge through physical coupling and stigmergic coordination when many robots operate in shared environments. Many of these principles are inspired by biological systems\, including our studies of construction and aggregation in honeybees and subterranean and mound-building termites. In this talk\, I will present examples from our lab\, including soft robots that exploit viscous fluid–structure interactions for articulated control\, microrobots that leverage magnetic and hydrodynamic interactions to produce a range of collective behaviors\, and entangled robotic matter that achieves cohesive motion through transient physical entanglement. Together\, these systems illustrate how intelligence can be distributed across morphology\, interactions\, and shared substrates\, enabling scalable and robust robot collectives. \n11:45 am–12:00 pm: Discussion \n12:00–1:30 pm: Catered Lunch \n1:30–4:30 pm: Discussion \n 
URL:https://cmsa.fas.harvard.edu/event/bioshape2_2026/
LOCATION:CMSA 20 Garden Street Cambridge\, Massachusetts 02138 United States
CATEGORIES:Programs
ATTACH;FMTTYPE=image/jpeg:https://cmsa.fas.harvard.edu/media/Biology2.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20260427T150000
DTEND;TZID=America/New_York:20260427T160000
DTSTAMP:20260410T174150
CREATED:20260210T203936Z
LAST-MODIFIED:20260210T203936Z
UID:10003899-1777302000-1777305600@cmsa.fas.harvard.edu
SUMMARY:Quantum Field Theory and Physical Mathematics
DESCRIPTION:Quantum Field Theory and Physical Mathematics Seminar \nSpeaker: Charles Young\, University of Hertfordshire
URL:https://cmsa.fas.harvard.edu/event/qft_42726/
LOCATION:CMSA Room G10\, CMSA\, 20 Garden Street\, Cambridge\, MA\, 02138\, United States
CATEGORIES:Quantum Field Theory and Physical Mathematics
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20260427T163000
DTEND;TZID=America/New_York:20260427T173000
DTSTAMP:20260410T174150
CREATED:20260324T172426Z
LAST-MODIFIED:20260324T172426Z
UID:10003924-1777307400-1777311000@cmsa.fas.harvard.edu
SUMMARY:Colloquium
DESCRIPTION:Colloquium \nSpeaker: Ofer Feinerman\, Weizmann Institute of Science \n  \n 
URL:https://cmsa.fas.harvard.edu/event/colloquium-42726/
LOCATION:CMSA Room G10\, CMSA\, 20 Garden Street\, Cambridge\, MA\, 02138\, United States
CATEGORIES:Colloquium
END:VEVENT
END:VCALENDAR