Seminars
CILVR Seminars
-
CILVR Seminar: Understanding LLM Training
Date and time: Wednesday, October 23, 2024, 2PMLocation: ONLISpeaker: Angelica ChenWatch the recording here.
-
CILVR Seminar: Leveraging Large Datasets and LLMs to Improve Healthcare and Health Equity
Date and time: Wednesday, October 16, 2024, 2PMLocation: 60FA , RoomSpeaker: Irene ChenWatch the recording here.
-
CILVR Seminar: Generalization in diffusion models arises from geometry adaptive harmonic representations
Date and time: Wednesday, October 9, 2024, 2PMLocation: 60FA , RoomSpeaker: Zahra KadkhodaieHere, we show that two DNNs trained on non-overlapping subsets of a dataset learn nearly the same score function, and thus the same density, when the number of training images is large enough.
Watch the recording here. -
Quantifying and Improving Generalization through Compression Bounds and Model Merging
Date and time: Wednesday, October 2, 2024, 2PMLocation: 60FA , Room 7th floor open spaceSpeaker: Sanae LotfiWatch the recording here.
-
CILVR SEMINAR: Neural Language Representations and Scaling Semi-Supervised Learning for Speech Recognition
Date and time: Thursday, April 25, 2024, 11AMLocation: 60FA , Room 7th floor open space, CDSSpeaker: Cal Peyser -
CILVR SEMINAR: Show Your Work with Confidence: Confidence Bands for Tuning Curves | Representations of Neural Network Training Dynamics
Date and time: Thursday, April 11, 2024, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Nick Lourie, Michael Hu -
CILVR SEMINAR: A recurrent network model of planning predicts hippocampal replay and human behavior
Date and time: Thursday, March 28, 2024, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Marcelo Mattar -
CILVR SEMINAR: A Framework for Multi-modal learning: Jointly Modeling Inter- and Intra-Modality Dependencies | Variance-Covariance Regularization Improves Representation Learning
Date and time: Thursday, March 14, 2024, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Divyam Madaan, Jiachen Zhu -
CILVR SEMINAR: Theory and Practice in Language Model Fine-Tuning
Date and time: Thursday, February 29, 2024, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Sadhika Malladi -
CILVR SEMINAR: Some Emerging Challenges in AI and Law
Date and time: Thursday, February 15, 2024, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Prof. Sunoo ParkThis talk will give an overview of a range of emerging challenges at the intersection of AI and law, including some ongoing copyright lawsuits, legal controversies around algorithmic decision making, and a brief overview of major regulatory efforts. (The talk will be aimed at computer scientists; no law background is required.)
-
CILVR SEMINAR: From the Sun to Birds: Using Computer Vision to Measure The Universe
Date and time: Thursday, February 1, 2024, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: David FouheyIn this talk, Prof. Fouhey will show off the systems that he built with members of his research group and collaborators. He will focus primarily on solar physics, where they have developed a series of systems with a primary focus on obtaining better measurements of the Sun's vector magnetic field.
Click here to view the recording.
-
CILVR SEMINAR: On Bringing Robots Home | NetHack is Hard to Hack
Date and time: Thursday, December 7, 2023, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Nur Muhammad “Mahi” Shafiullah, Ulyana PiterbargIn this talk Nur Muhammad “Mahi” Shafiullah will talk on the topic "On Bringing Robots Home" and Ulyana Piterbarg will talk on the topic "NetHack is Hard to Hack".
Click here to view the recording
-
CILVR SEMINAR: Discrete Generative Models for Designing and Sampling Atomistic Data | Practical causal representation learning with an asymmetric prior
Date and time: Monday, November 13, 2023, 8:17PMLocation: 60FA , Room 7th floor common areaSpeaker: Nate Gruver, Taro MakinoIn this talk, Nate Gurver will discuss two of his recent projects on applying generative sequence models to atomistic data, in particular protein design and discovery of stable inorganic materials. Taro Makino will talk about Asymmetric Prior Variational Autoencoder (AP-VAE)
Click here to view the recording
-
CILVR SEMINAR: World Knowledge in the Time of Large Models
Date and time: Thursday, November 2, 2023, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Kenneth MarinoThis talk will discuss the massive shift that has come about in the vision and ML community as a result of the large pre-trained language and language and vision models such as Flamingo, GPT-4, and other models.
Click here to view the recording.
-
CILVR SEMINAR: Self-Supervised Learning- Going beyond images
Date and time: Thursday, October 26, 2023, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Quentin GarridoIn this talk, we will dive into self-supervised learning (SSL) and how to generalise ideas developed for images to other domains.
Click here to view the recording
-
CILVR SEMINAR: Objective-Driven AI- towards AI systems that can learn, remember, reason, and plan
Date and time: Thursday, October 19, 2023, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Yann LeCun -
CILVR SEMINAR: Large learning rate in matrix recovery problems | Data-driven multiscale modeling of subgrid parameterizations in climate models
Date and time: Thursday, October 5, 2023, 11AMLocation: 60FA , Room 7th floor common areaSpeaker: Lei Chen and Karl OtnessSubgrid parameterizations, which represent physical processes occurring below the resolution of current climate models, are an important component in producing accurate, long-term predictions for the climate. In this talk Karl will present this research problem and discuss an ongoing project investigating a multiscale, deep learning-based approach to this task. Lei will introduce some of his ongoing research on optimization instability and implicit bias.
-
CILVR SEMINAR: Novel Loss, Layer and Applications for Neural Network based Continual Learning
Date and time: Thursday, September 28, 2023, 11AMLocation: 60FA , Room 7th Floor Open SpaceSpeaker: Sungmin ChaSungmin will introduce some of his past research on continual learning using neural networks. To do so, he will briefly explain the settings of continual learning research and then discuss his research findings related to the novel loss function, layer, and applications for continual learning. Finally, he will briefly discuss the limitations and future works.
Click here to view the recording
-
CILVR SEMINAR: Structured Models of Music
Date and time: Tuesday, May 17, 2022, 8PMLocation: ONLISpeaker: Jesse Engel, Google Magenta -
CILVR SEMINAR: Towards Language Technology for a Truly Multilingual World?
Date and time: Tuesday, May 3, 2022, 8PMLocation: ONLISpeaker: Ivan Vulic, University of CambridgeLanguage technology tools such as Google Translate or virtual assistants (Siri, Alexa) were components of collective SciFi-inspired imagination not many years ago. Today, they are an essential driver of the digital AI transformation, used by hundreds of millions of people. I will introduce a range of recent techniques, breakthroughs and lessons learned that aim to deal with large cross-language variations and low-data learning regimes.
-
CILVR SEMINAR: Towards Real-World Meta-Learning
Date and time: Tuesday, April 19, 2022, 8PMLocation: ONLISpeaker: Hae Beom Lee, KAISTI will introduce my previous efforts extending the scope of meta-learning towards real-world learning scenarios. We will see how to deal with realistic task distributions including class/task imbalance and distributional shift, discuss how to develop practical meta-knowledge applicable to arbitrary CNN architectures, and also how to scale up few-shot learning to many-shot and heterogeneous learning problems.
-
CILVR SEMINAR: Visual Learning in the Open World
Date and time: Tuesday, April 5, 2022, 8PMLocation: ONLISpeaker: Mengye Ren, Google Brain (Toronto)We have seen machine learning making great strides in understanding visual scenes. Yet, most of its success relies on training models on a massive amount of data in a closed world and evaluating them in a similar environment. I present an alternative paradigm that will allow machines to acquire visual knowledge through an online stream of data in an open world, which entails abilities such as learning visual representations and concepts efficiently with limited and non-iid data.
-
CILVR SEMINAR: Taming Large Pre-Trained Neural Language Models: Differentiable Game-Theoretic Regularization and Sensitivity-Guided Optimization
Date and time: Tuesday, March 29, 2022, 8PMLocation: ONLISpeaker: Tuo Zhao, Georgia TechPre-trained language models have fundamentally changed the landscape of NLP. However, as the pre-trained language models are becoming increasingly large, we have also witnessed that the gain in their generalization performance is becoming marginal, especially when we only have limited labelled data in downstream tasks. To improve their generalization, we propose a new framework for fine-tuning of pretrained models to yield better generalization performance.
-
CILVR SEMINAR: Deep learning for scientific discovery
Date and time: Tuesday, March 8, 2022, 7:43PMLocation: ONLISpeaker: Carlos Fernandez-GrandaDeep-learning models for image processing achieve impressive results when trained on standard natural-image datasets in a supervised fashion. However, unleashing their potential in practice will require developing unsupervised or semi-supervised approaches capable of learning from real data, as well as understanding the strategies learned by these models. In this talk, we will describe advances in this direction.
-
CILVR SEMINAR: The Lottery Ticket Hypothesis: On Sparse, Trainable Neural Networks
Date and time: Tuesday, February 22, 2022, 9PMLocation: ONLISpeaker: Jonathan Frankle, MITI recently proposed the lottery ticket hypothesis: that the dense neural networks we typically train have much smaller subnetworks capable of reaching full accuracy from early in training. In this talk, I will discuss established results and the latest developments in my line of work.
-
CILVR SEMINAR: Foundations behind the Prescient Design’s approach to protein design
Date and time: Tuesday, February 8, 2022, 9PMLocation: ONLISpeaker: Kyunghyun ChoPrescient Design at Genentech investigates, develops and deploys a machine learning driven platform for protein design. At the core of Prescient Design’s strategy is the deep manifold sampler which is a non-autoregressive sequence denoising autoencoder combined with a function predictor. In this talk, I will describe the foundations behind the deep manifold sampler.
-
CILVR SEMINAR: Characterizing and Mitigating Threats to Trust and Safety Online
Date and time: Wednesday, December 15, 2021, 8PMLocation: ONLISpeaker: Yiqing Hua, Cornell TechSupporting a safe and trustworthy online environment is challenging, as these environments are constantly threatened by abusive behaviors that cause real human harm. I will present my work on characterizing threats, and empowering users with new techniques to combat these threats.
-
CILVR SEMINAR: Student Research
Date and time: Wednesday, December 1, 2021, 8PMLocation: ONLISpeaker: VariousJiwoong Im: Causal Effect Variational Autoencoder with Uniform Treatment
Aahlad Puli: Predictive Modeling in the Presence of Nuisance-Induced Spurious Correlations
-
CILVR SEMINAR: Deep learning from an information theory perspective
Date and time: Wednesday, November 17, 2021, 8PMLocation: ONLISpeaker: Ravid Shwartz-ZivWhile DNNs have achieved many breakthroughs, our understanding of their internal structure, optimization process, and generalization is poor, and we often treat them as black boxes. We attempt to resolve these issues by suggesting that DNNs learn to optimize the Information Bottleneck (IB) principle - the tradeoff between information compression and prediction quality.
-
CILVR SEMINAR: Understanding Language Referring to the Visual World
Date and time: Wednesday, November 3, 2021, 7PMLocation: ONLISpeaker: Volkan Cirik, CMUCurrent AI systems have limited agency in the world, i.e., they lack the embodiment and sensory experience of the world. Hopefully, embodied AI systems will address these issues, such as autonomous vehicles mobilizing the visually impaired in our communities. The manifestation of AI in the physical world necessitates research into linking natural language to physical referents. The central theme of this talk is to understand and model this linking process, namely language grounding.
-
CILVR Seminar: Learning to predict fewer degenerate sequences using autoregressive language models
Date and time: Wednesday, October 20, 2021, 7PMLocation: ONLISpeaker: Ilia KulikovUnlikelihood training and the consistency of distributions induced by language models and decoding algorithms.
-
CILVR SEMINAR: Update on Student Research
Date and time: Wednesday, September 22, 2021, 7PMLocation: ONLISpeaker: Students- Min Jae: On the Cryptographic Hardness of Learning Neural Networks.
- Mahi Shaffiullah: Surprising Effectiveness of Representation Learning for Visual Imitation.
- Ben Evans: Context is Everything: Implicit Identification for Dynamics Adaptation.
- Nan Wu: Joint speech and text pre-training for text normalization towards end-to-end TTS.
-
CILVR SEMINAR: Faculty Research
Date and time: Wednesday, September 8, 2021, 7PMLocation: ONLISpeaker: VariousTal Linzen, Sam Bowman, Lerrel Pinto, and Jason Weston (Meta) will present their recent research.
-
CILVR SEMINAR : How to Build an AI-Based Medical Device
Date and time: Monday, April 5, 2021, 5PMLocation: ONLISpeaker: Sumit Chopra -
CILVR SEMINAR : Distribution-Free, Risk-Controlling Prediction Sets
Date and time: Sunday, March 21, 2021, 5PMLocation: ONLISpeaker: Stephen Bates -
CILVR SEMINAR :From ``What'' to ``Why'': Towards Causal Deep Networks
Date and time: Sunday, March 7, 2021, 6PMLocation: ONLISpeaker: Rosemary Ke -
CILVR SEMINAR : Learning to Plan and Planning to Learn
Date and time: Sunday, February 21, 2021, 6PMLocation: ONLISpeaker: Aviv Tamar -
CILVR SEMINAR : On Fine-Tuning of Pretrained Language Models under Limited or Weak Supervision
Date and time: Sunday, February 7, 2021, 6PMLocation: ONLISpeaker: Tuo Zhao