May 2024 - talks - lists.cs.princeton.edu

Karhan Kayan will present his General Exam "Infinigen Indoors: Photorealistic Indoor Scenes using Procedural Generation" on Tuesday, May 7, 2024 at 3:30 PM in CS 301.
by CS Grad Department 02 May '24

02 May '24

Karhan Kayan will present his General Exam " Infinigen Indoors: Photorealistic Indoor Scenes using Procedural Generation " on Tuesday, May 7, 2024 at 3:30 PM in CS 301 . Committee Members: Jia Deng (advisor), Szymon Rusinkiewicz , Adam Finkelstein Abstract: We introduce Infinigen Indoors, a Blender-based procedural generator of photorealistic indoor scenes. It builds upon the existing Infinigen system, which focuses on natural scenes, but expands its coverage to indoor scenes by introducing a diverse library of procedural indoor assets, including furniture, architecture elements, appliances, and other day-to-day objects. It also introduces a constraint-based arrangement system, which consists of a domain-specific language for expressing diverse constraints on scene composition, and a solver that generates scene compositions that maximally satisfy the constraints. We provide an export tool that allows the generated 3D objects and scenes to be directly used for training embodied agents in real-time simulators such as Omniverse and Unreal. Infinigen Indoors will be open-sourced under the BSD license. Reading List: https://docs.google.com/document/d/1MVaGLJx5293qyQZJN1PQLa1QJBHnfBV6TzzTBYJ… Everyone is invited to attend the talk, and those faculty wishing to remain for the oral exam following are welcome to do so.

1 0

Alexander Strzalkowski will present his FPO "Inferring the Biological Time of Single Cells using Supervised Dimensionality Reduction and Trees" on Monday, May 6, 2024 at 11:00 AM in CS 301.
by gradinfo＠cs.princeton.edu 02 May '24

02 May '24

Alexander Strzalkowski will present his FPO "Inferring the Biological Time of Single Cells using Supervised Dimensionality Reduction and Trees" on Monday, May 6, 2024 at 11:00 AM in CS 301. Location: CS 301 The members of Alexander's committee are as follows: Examiners: Ben Raphael (Adviser), Yuri Pritykin, Michelle Chan Readers: Ellen Zhong, Li Ding (Washington University in St. Louis) Everyone is invited to attend his talk. Abstract follows below: Single-cell omics measurements have exploded in growth over the past decade. This explosion has allowed researchers to probe human health and biology with unprecedented resolution. Currently, all these types of measurements are destructive, thus they only provide static snapshots of important dynamic biological processes such as development, cancer progression, and cell cycle. As cells differentiate/progress biologically asynchronously in most tissues, a major computational task often called trajectory inference is to infer the latent biological time also known as pseudotime of every cell. This inverse problem in general is quite challenging and is further complicated by the fact that single-cell omics measurements like scRNA-seq and scATAC-seq are highly sparse and highdimensional. Much of my work has shown that simple linear supervised dimensionality reduction techniques that rely on cell type information can outperform complex non-linear dimensionality reduction techniques when used in conjunction with state-of-the-art trajectory inference methods in a large benchmark. Moreover, we investigate the difficulties of benchmarking trajectory inference methods in the absence of ground truth showcasing that the implicit goal of many methods is not to identify intermediate/transient cell types but rather order cell types. In addition, we introduce a novel supervised linear dimensionality reduction technique called BCA that when applied to simulated and real datasets is better able to uncover intermediate cell types. Lastly, we have been interested in modeling the relationship between cell lineages of inferred phylogenies from single-cell lineage tracing data and scRNA-seq trajectories (the partial ordering of cells induced by pseudotimes). We have found that by using a novel irreversible continuous state model of pseudotime on a rooted tree that we are better able to model unobserved ancestral pseudotimes in simulated and real phylogenies.

1 0

Lingjie Mei will present his General Exam "Procedurally Generating 3D Indoor Scenes for Computer Vision and Robotics" on Monday, May 6, 2024 at 3:00 PM in Friend 109 and via zoom.
by CS Grad Department 01 May '24

01 May '24

Lingjie Mei will present his General Exam "Procedurally Generating 3D Indoor Scenes for Computer Vision and Robotics" on Monday, May 6, 2024 at 3:00 PM in Friend 109 and via zoom. Zoom link: https://princeton.zoom.us/j/8415068766 Committee Members: Jia Deng (advisor), Felix Heide, Ellen Zhong Abstract: Synthetic data rendered by conventional computer graphics has seen increasing adoption in computer vision and AI research, especially for 3D vision and embodied AI. Synthetic data can be rendered in unlimited quantities and can automatically provide high-quality 3D ground truth, enabling large-scale training of computer vision models and embodied agents. We propose a synthetic data generator that can produce infinite photorealistic 3D indoor scenes. Our system is entirely procedural: every asset, including furniture, architecture elements, appliances, and other day-to-day objects, is generated from scratch via mathematical rules and randomized parameters. We also introduce a constraint-based arrangement system, which consists of a domain-specific language for expressing diverse constraints and preferences on indoor scene composition, and a solver that generates floorplans and scene layouts that maximally satisfy the constraints. The scenes are photorealistic, generated with pixel-perfect annotations, and can be exported to real-time simulators and run at interactive frame rates. Overall, anyone can directly use and customize our system to generate infinite high-quality indoor scene data for a variety of computer vision tasks, including object detection, semantic segmentation, optical flow, and 3D reconstruction. Robot scientists can also use the data in simulation for manipulation and grasping. We expect this to be a useful tool for researchers beyond. Reading List: https://docs.google.com/document/d/1mmf5W0sKZkLgWVKXpCA7TkSSWUfBoeuKFXWHw8k… Everyone is invited to attend the talk, and those faculty wishing to remain for the oral exam following are welcome to do so.

1 0

Stephen Newman will present his General Exam "Decentralization Cheapens Corruptive Majority Attacks" on May 8, 2024 at 1:30pm in Friend 005.
by Nicki Mahler 01 May '24

01 May '24

Stephen Newman will present his General Exam "Decentralization Cheapens Corruptive Majority Attacks" on May 8, 2024 at 1:30pm in Friend 005. Committee members: Matt Weinberg (adviser), Mark Braverman, and Huacheng Yu Abstract and Reading List: [ https://docs.google.com/document/d/1JBsVhPuw0e5nqK3dv7EbvpfnnfQ4-5ga2qFy_HN… | https://docs.google.com/document/d/1JBsVhPuw0e5nqK3dv7EbvpfnnfQ4-5ga2qFy_HN… ] Abstract: Corruptive majority attacks, in which mining power is distributed among miners and an attacker attempts to bribe a majority of miners into participation in a majority attack, pose a threat to blockchains. Budish bounded the cost of bribing miners to participate in an attack by their expected loss as a result of attack success. We show that this bound is loose. In particular, an attack may be structured so that under equilibrium play by most miners, a miner's choice to participate only slightly affects the attack success chance. Combined with the fact that most of the cost of attack success is externalized by any given small miner, this implies that if most mining power is controlled by small miners, bribing miners to participate in such an attack is much cheaper than the Budish bound. We provide a scheme for a cheap corruptive majority attack and discuss practical concerns and consequences.

1 0

Andre Niyongabo Rubungo will present his General Exam "Leveraging Large Language Models for Materials Science" on Monday, May 6, 2024 at 9:00 AM in FC 005.
by CS Grad Department 01 May '24

01 May '24

Andre Niyongabo Rubungo will present his General Exam "Leveraging Large Language Models for Materials Science" on Monday, May 6, 2024 at 9:00 AM in FC 005. Committee Members: Adji Bousso Dieng (advisor), Danqi Chen, Tom Griffiths Abstract: Predicting material properties plays an important role in materials design and discovery. Having a robust and accurate materials property predictor can help in screening materials of interest and in designing new materials with desired properties. Before the machine learning era, ab initio methods such as density functional theory (DFT) that allow the prediction and calculation of material behavior on the basis of quantum mechanical considerations were used to facilitate this process. DFT methods are somewhat accurate, however they are extremely slow, very computationally expensive, and hard to scale. Recently, scientists have started leveraging machine learning as an alternative to DFT. More specifically, graph neural networks (GNNs) have been particularly effective in capturing the interactions in materials for the purpose of property prediction. Crystal lattice sites are represented as nodes and the bonds (e.g., ionic or covalent or van der Waals) between them are represented as edges. GNNs then typically learn the contextual representation of each node and edge within the crystal graph to predict its properties. Although GNNs are powerful, they still face several challenges when it comes to predicting crystal properties. For example, they fail to efficiently encode the periodicity inherent to any crystal which results from the repetitive arrangement of unit cells within a lattice. Furthermore, current state-of-the-art GNNs do not account for space group information which is critical for characterizing crystals. Finally, GNNs are known to suffer from a problem known as over-smoothing which may hinder expressivity. This research marks a pioneering step in leveraging the powerful learning capabilities of large language models (LLMs) to improve the process of discovering new materials. We aim to use LLMs to learn the representation of a crystal structure given either its chemical formula, structure strings, or structure text descriptions as input. Representing a crystal as text instead of a graph has several advantages. Textual data contain rich information and are very expressive; additionally, incorporating desired information in text is generally more straightforward compared to graphs. We intend to design and implement a robust and accurate LLM-based property predictor by first collecting enough labeled materials data and using it to finetune available powerful pretrained LLMs. Furthermore, by having an accurate LLM-based property predictor, we can finetune an LLM generative model that can generate multiple samples of structure strings or descriptions of new materials and use our LLM-based property predictor to predict the stability of the newly generated materials and hence only keep the most stable ones. A strategy that simulates the experimental approach to materials design and discovery, but in a faster and more efficient way. Reading List: https://docs.google.com/document/d/14bGQagXTIxpzgtg9dGsaZwYo4nm96xgI1pAoGLb… Everyone is invited to attend the talk, and those faculty wishing to remain for the oral exam following are welcome to do so.

1 0

Gongqi Huang will present his General Exam "Designing and Implementing Functional, Secure, and Correct Information Flow Control Interfaces" on May 3, 2024 at 1:30pm in Friend 007.
by Nicki Mahler 01 May '24

01 May '24

Gongqi Huang will present his General Exam "Designing and Implementing Functional, Secure, and Correct Information Flow Control Interfaces" on May 3, 2024 at 1:30pm in Friend 007. The members of his committee are as follows: Amit Levy(adviser), Wyatt Lloyd, and Margaret Martonosi Abstract and Reading list can be found here: [ https://docs.google.com/document/d/1aUcos6drI5fzcYxu0YwkHX2GZ6iEdmJ8c_EEduB… | https://docs.google.com/document/d/1aUcos6drI5fzcYxu0YwkHX2GZ6iEdmJ8c_EEduB… ] [ https://docs.google.com/document/d/1990kg2Qtj24X8sjkH1Cq5dA_0hFwIDFj0gCLuzX… | https://docs.google.com/document/d/1990kg2Qtj24X8sjkH1Cq5dA_0hFwIDFj0gCLuzX… ]

1 0