We will demonstrate how to make a personalized AI partner that can do certain tasks for you. This step by step process can be used to develop your own companion.

Understand what is a Gemini Gems and how it can be personalized for your custom needs. We will show you how to capture the tasks that you would like to complete and give you higher quality responses that will ensure your communication and tasks are completed with ease.

Register here.

Abstract: To be helpful assistants, AI agents must be aware of their own capabilitiesand limitations. This includes knowing when to answer from parametricknowledge versus using tools, when to trust tool outputs, and when toabstain or hedge. Such capabilities are hard to teach through supervisedfine-tuning because they require constructing examples that reflect theagent's specific capabilities. We therefore propose a radically new approachto teaching agents what they know: collaborative self-play. We constructmulti-agent collaborations in which the group is rewarded for collectivelyarriving at correct answers. The desired meta-knowledge emerges fromthe incentives built into the structure of the interaction. We focus on smallsocieties of agents that have access to heterogeneous tools (corpus-specificretrieval), and therefore must collaborate to maximize their success withminimal effort. Experiments show that group-level rewards for multiagent communities can induce policies that transfer to improve tool use andselective prediction in single-agent scenarios. Speaker: Syed Mostofa Monsur Location: Old Computer Science Building - CS2311

You are cordially invited to attend the biweekly Brookhaven AI Mixer (BAM). BAM includes one short talk on AI research happening at BNL, followed by an open mixer over coffee and snacks for everyone to network and discuss all things AI. The first half hour will consist of presentations that will be available via ZOOM, and the second half hour will be for in person only networking.

Join us every other Tuesday at noon in CDSD's Training Room (building 725, 2nd floor) to learn about interesting AI methods and applications, engage with potential collaborators, prepare for pending FASST funding calls, and build a community of AI for Science at BNL.

Learning Generalizable Program and Architecture Representations for Performance Modeling

Abstract: Performance modeling is an essential tool in many areas of computer science and engineering. However, existing performance modeling approaches have limitations, such as high computational cost, narrow flexibility, or restricted accuracy/generality. To address these limitations, this talk introduces PerfVec, a novel deep learning-based performance modeling framework that learns high-dimensional and independent/orthogonal program and microarchitecture representations. Once learned, a program representation can be used to predict its performance on any microarchitecture, and likewise, a microarchitecture representation can be applied in the performance prediction of any program. Additionally, PerfVec yields a foundation model that captures the performance essence of instructions, which can be directly used by developers in numerous performance modeling-related tasks without incurring its training cost. The evaluation demonstrates that PerfVec is more general and efficient than previous approaches. This talk will also introduce how PerfVec's design principles can benefit broader research areas.

Biography: Lingda Li is a computer scientist at Brookhaven National Laboratory. He is generally interested in computer architecture and programming model research, with focus on simulation/modeling, memory systems, and machine learning. Before joining BNL, he worked at the Department of Computer Science of Rutgers University as a postdoc to carry out GPGPU research. He obtained a PhD in computer architecture from the Microprocessor Research and Development Center at Peking University.

Location: CDS, Bldg. 725, Training Room

Join ZoomGov Meeting: https://bnl.zoomgov.com/j/1605837856?pwd=kYqJs4bVBt4E0cMCWR6GXH3wxzOoiw.1

Meeting ID: 160 583 7856
Passcode: 161580

Abstract:

Conventional approaches to scientific discovery often prioritize building larger sensors, gathering more data, and scaling up computational power. In this talk, I will present a complementary perspective: extracting insights hidden in the data we already have. The key lies in using AI not as a black-box predictor, but as a tool for interpreting data through its underlying physical process.

I will demonstrate how AI, when integrated with the physics of light propagation, can serve as a computational lens to overcome fundamental limitations in fields ranging from biomedicine to astrophysics. Specifically, I will showcase two compelling applications: non-invasive imaging through scattering biological tissues, and detecting faint exoplanets against the overwhelming brightness of their host stars.

These methods represent a departure from traditional learning-based approaches that rely on fitting models to training labels and hoping for generalization. Instead, with physics-informed strategies that decode how light propagates, we can transform raw measurements into scientifically meaningful insights--without requiring costly hardware upgrades or human-annotated datasets. Finally, I will outline future directions for combining AI with physical principles, enabling us to unlock more phenomena once considered hidden and accelerating discoveries in healthcare, astronomy, and beyond.

Short Bio:

Brandon Y. Feng is a Postdoctoral Associate at MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) and a Visiting Scientist at the Harvard-Smithsonian Center for Astrophysics. His research bridges artificial intelligence and physics to expand the limits of human and machine vision. He develops AI-driven methods that reveal hidden patterns in complex visual data, driving breakthroughs in areas such as exoplanet detection and imaging through scattering tissues. His work has been published in top venues, including Science Advances, CVPR, ICCV, ECCV, and NeurIPS, and has been featured in Science.org, New Scientist, and Phys.org. He holds a Ph.D. in Computer Science from the University of Maryland, along with a B.A. in Computer Science and Statistics and an M.S. in Statistics from the University of Virginia.

Location: NCS 220

Over the past decade, researchers in neuroscience, psychology and artificial intelligence have come together to build advanced computer models that mimic how our brain processes what we see. These models are designed to closely copy the brain's visual system, all the way to a key area called the inferior temporal cortex, which plays an important role in recognizing objects.

Because these computer models can be fully observed, scientists can use them to make detailed predictions about how the brain works -- something older, more theoretical models could not do.

Dr. James DiCarlo's work explores whether these computer digital twin models of the brain could help guide safe, non- invasive ways to infl uence brain activity. In his talk, he explains how such a model could be used to design specific patterns of light. When this carefully designed light is added to what the eye naturally sees, it can precisely influence activity in groups of neurons in the inferior temporal cortex.

Since neural activity in this visual brain area may be connected to emotional states like anxiety, this research could eventually open the door to non-invasive approaches that may benefit mental well-being in the future.

Speaker: James J. DiCarlo, MD, PhD, Peter de Florez Professor, MIT Brain and Cognitive Sciences, and Director, MIT Siegel Family Quest for Intelligence

Location: Staller Center Main Stage

The event will be livestreamed at stonybrook.edu/live

How Language Makes us Smart (without Big Data) presented by Charles Yang

Abstract: Language provides the glue that combines simpler concepts into complex ones. To study how language guides conceptual development, we need precise accounts of how rules are learned from the child's linguistic experience, which is extremely limited in comparison to the amount of data available to current machine learning methods. In this talk, I discuss a mathematical model of inductive generalization, which enables language learning with very small amount of data. Such a view of learning has strong implications for the cross-cultural/linguistic variation of development. As a case study, I show that Hong Kong children learning Cantonese, which has a relatively simpler formal counting system, develop understanding of symbolic numbers a full year ahead of English-learning children in the United States, which is precisely predictable from the learning model. The new conception of learning adds another wrinkle to the eternal question of how language and thought are related to each other.

Bio: Charles Yang studied at the MIT AI lab and now teaches linguistics, computer science and psychology and directs the Program in Cognitive Science at the University of Pennsylvania. He is the author of several books: The Price of Linguistic Productivity (2016 MIT Press) won the Leonard Bloomfield Award from the Linguistic Society of America. His honors include a Guggenheim fellowship.
ICB&DD 19th Annual Symposium

Iwao Ojima, Director, ICB&DD
Ivet Bahar Chair, Organizing Committee
Dima KozakovCo-Chair, OrganizingCommittee

There will be poster sessions on projects conducted in the ICB&DD member's laboratories aswell as other laboratories in the area. Awards will be given to the best three posters.

Please see the link for the registration and poster sessions in:
https://www.stonybrook.edu/commcms/icbdd/https://forms.gle/Wh4UzVx9U4HWStXb8
Abstract: Reward hacking, where a reasoning model exploits loopholes in a reward function to achieve high rewards without solving the intended task, poses a significant threat. This behavior may be explicit, i.e. verbalized in the model's chain-of-thought (CoT), or implicit, where the CoT appears benign thus bypasses CoT monitors. To detect implicit reward hacking, we propose TRACE (Truncated Reasoning AUC Evaluation). Our key observation is that hacking occurs when exploiting the loophole is easier than solving the actual task. This means that the model is using less effort than required to achieve high reward. TRACE quantifies effort by measuring how early a model's reasoning becomes sufficient to obtain the reward. We progressively truncate a model's CoT at various lengths, force the model to answer, and estimate the expected reward at each cut-off. A hacking model, which takes a shortcut, will achieve a high expected reward with only a small fraction of its CoT, yielding a large area under the reward-vs-length curve. TRACE achieves over 65% gains over our strongest 72B CoT monitor in math reasoning, and over 30% gains over a 32B monitor in coding. We further show that TRACE can discover unknown loopholes during training. Overall, TRACE offers a scalable unsupervised approach for oversight where current monitoring methods prove ineffective.

Speaker: Dikshya

Location: Old Computer Science Building - CS2311
Ready to stop taking notes and make your virtual meetings work for you? Join our upcoming workshop to master Zoom's built-in AI tools and streamline your workday. You'll learn how to generate instant AI Meeting Summaries to capture key action items effortlessly, use live in-meeting Q&A to ask Zoom AI questions and catch up on missed details in real time, and prompt Zoom to build sleek, professional virtual backgrounds on the fly. Please click the link below to register. We hope to see you there!

https://stonybrookuniversity.co1.qualtrics.com/jfe/form/SV_a9PVlBw0E1Bal1A?source=campus%20announcements&session=100910261600zoom
AI Seminar: Video Architecture Search - Michael Ryoo Abstract: Video understanding is a challenging problem. Because a video contains spatio-temporal data, its feature representation is required to abstract both appearance and motion information. This is not only essential for automated understanding of the semantic content of videos, such as Web-video classification or sport activity recognition, but is also crucial for robot perception and learning. Previously, convolutional neural networks (CNNs) for videos were normally built by manually extending known 2D architectures such as Inception and ResNet to 3D or by carefully designing two-stream CNN architectures that fuse together both appearance and motion information. However, designing an optimal video architecture to best take advantage of spatio-temporal information in videos still remains an open problem. In this talk, we discuss recent progress in neural architecture search for videos, obtaining more optimal network architectures for video understanding.