Deep Policy Gradient Methods Without Batch Updates, Target Networks, or Replay Buffers

Autor: Vasan, Gautham, Elsayed, Mohamed, Azimi, Alireza, He, Jiamin, Shariar, Fahim, Bellinger, Colin, White, Martha, Mahmood, A. Rupam

Modern deep policy gradient methods achieve effective performance on simulated robotic tasks, but they all require large replay buffers or expensive batch updates, or both, making them incompatible for real systems with resource-limited computers. We

Externí odkaz: http://arxiv.org/abs/2411.15370

Design-o-meter: Towards Evaluating and Refining Graphic Designs

Autor: Goyal, Sahil, Mahajan, Abhinav, Mishra, Swasti, Udhayanan, Prateksha, Shukla, Tripti, Joseph, K J, Srinivasan, Balaji Vasan

Graphic designs are an effective medium for visual communication. They range from greeting cards to corporate flyers and beyond. Off-late, machine learning techniques are able to generate such designs, which accelerates the rate of content production

Externí odkaz: http://arxiv.org/abs/2411.14959

Reflections from the 2024 Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry

Autor: Zimmermann, Yoel, Bazgir, Adib, Afzal, Zartashia, Agbere, Fariha, Ai, Qianxiang, Alampara, Nawaf, Al-Feghali, Alexander, Ansari, Mehrad, Antypov, Dmytro, Aswad, Amro, Bai, Jiaru, Baibakova, Viktoriia, Biswajeet, Devi Dutta, Bitzek, Erik, Bocarsly, Joshua D., Borisova, Anna, Bran, Andres M, Brinson, L. Catherine, Calderon, Marcel Moran, Canalicchio, Alessandro, Chen, Victor, Chiang, Yuan, Circi, Defne, Charmes, Benjamin, Chaudhary, Vikrant, Chen, Zizhang, Chiu, Min-Hsueh, Clymo, Judith, Dabhadkar, Kedar, Daelman, Nathan, Datar, Archit, Evans, Matthew L., Fard, Maryam Ghazizade, Fisicaro, Giuseppe, Gangan, Abhijeet Sadashiv, George, Janine, Gonzalez, Jose D. Cojal, Götte, Michael, Gupta, Ankur K., Harb, Hassan, Hong, Pengyu, Ibrahim, Abdelrahman, Ilyas, Ahmed, Imran, Alishba, Ishimwe, Kevin, Issa, Ramsey, Jablonka, Kevin Maik, Jones, Colin, Josephson, Tyler R., Juhasz, Greg, Kapoor, Sarthak, Kang, Rongda, Khalighinejad, Ghazal, Khan, Sartaaj, Klawohn, Sascha, Kuman, Suneel, Ladines, Alvin Noe, Leang, Sarom, Lederbauer, Magdalena, Liao, Sheng-Lun Mark, Liu, Hao, Liu, Xuefeng, Lo, Stanley, Madireddy, Sandeep, Maharana, Piyush Ranjan, Maheshwari, Shagun, Mahjoubi, Soroush, Márquez, José A., Mills, Rob, Mohanty, Trupti, Mohr, Bernadette, Moosavi, Seyed Mohamad, Moßhammer, Alexander, Naghdi, Amirhossein D., Naik, Aakash, Narykov, Oleksandr, Näsström, Hampus, Nguyen, Xuan Vu, Ni, Xinyi, O'Connor, Dana, Olayiwola, Teslim, Ottomano, Federico, Ozhan, Aleyna Beste, Pagel, Sebastian, Parida, Chiku, Park, Jaehee, Patel, Vraj, Patyukova, Elena, Petersen, Martin Hoffmann, Pinto, Luis, Pizarro, José M., Plessers, Dieter, Pradhan, Tapashree, Pratiush, Utkarsh, Puli, Charishma, Qin, Andrew, Rajabi, Mahyar, Ricci, Francesco, Risch, Elliot, Ríos-García, Martiño, Roy, Aritra, Rug, Tehseen, Sayeed, Hasan M, Scheidgen, Markus, Schilling-Wilhelmi, Mara, Schloz, Marcel, Schöppach, Fabian, Schumann, Julia, Schwaller, Philippe, Schwarting, Marcus, Sharlin, Samiha, Shen, Kevin, Shi, Jiale, Si, Pradip, D'Souza, Jennifer, Sparks, Taylor, Sudhakar, Suraj, Talirz, Leopold, Tang, Dandan, Taran, Olga, Terboven, Carla, Tropin, Mark, Tsymbal, Anastasiia, Ueltzen, Katharina, Unzueta, Pablo Andres, Vasan, Archit, Vinchurkar, Tirtha, Vo, Trung, Vogel, Gabriel, Völker, Christoph, Weinreich, Jan, Yang, Faradawn, Zaki, Mohd, Zhang, Chi, Zhang, Sylvester, Zhang, Weijie, Zhu, Ruijie, Zhu, Shang, Janssen, Jan, Foster, Ian, Blaiszik, Ben

Here, we present the outcomes from the second Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry, which engaged participants across global hybrid locations, resulting in 34 team submissions. The submissions spann

Externí odkaz: http://arxiv.org/abs/2411.15221

Test-time Conditional Text-to-Image Synthesis Using Diffusion Models

Autor: Shukla, Tripti, Karanam, Srikrishna, Srinivasan, Balaji Vasan

We consider the problem of conditional text-to-image synthesis with diffusion models. Most recent works need to either finetune specific parts of the base diffusion model or introduce new trainable parameters, leading to deployment inflexibility due

Externí odkaz: http://arxiv.org/abs/2411.10800

Analysis of Quantitative Angiography using Projection Foreshortening Correction and Injection Bias Removal

Autor: Mondal, Parmita, Shields, Allison, Bhurwani, Mohammad Mahdi Shiraz, Williams, Kyle A, Nagesh, Swetadri Vasan Setlur, Siddiqui, Adnan H, Ionita, Ciprian N

This study aims to mitigate these biases and enhance QA analysis by applying a path-length correction (PLC) correction, followed by singular value decomposition (SVD)-based deconvolution, to angiograms obtained through both in-silico and in-vitro met

Externí odkaz: http://arxiv.org/abs/2411.08185