Zobrazeno 1 - 10
of 17 352
pro vyhledávání: '"A De Sa"'
Autor:
Feng, Shuangquan, de Sa, Virginia R.
Automatic facial action unit (AU) recognition is used widely in facial expression analysis. Most existing AU recognition systems aim for cross-participant non-calibrated generalization (NCG) to unseen faces without further calibration. However, due t
Externí odkaz:
http://arxiv.org/abs/2409.00240
Autor:
de Sá, Alex G. C., Ascher, David B.
Machine learning (ML) is revolutionising drug discovery by expediting the prediction of small molecule properties essential for developing new drugs. These properties -- including absorption, distribution, metabolism and excretion (ADME)-- are crucia
Externí odkaz:
http://arxiv.org/abs/2408.00421
Autor:
Hartung, Michael, Maier, Andreas, Delgado-Chaves, Fernando, Burankova, Yuliya, Isaeva, Olga I., Patroni, Fábio Malta de Sá, He, Daniel, Shannon, Casey, Kaufmann, Katharina, Lohmann, Jens, Savchik, Alexey, Hartebrodt, Anne, Chervontseva, Zoe, Firoozbakht, Farzaneh, Probul, Niklas, Zotova, Evgenia, Tsoy, Olga, Blumenthal, David B., Ester, Martin, Laske, Tanja, Baumbach, Jan, Zolotareva, Olga
Most complex diseases, including cancer and non-malignant diseases like asthma, have distinct molecular subtypes that require distinct clinical approaches. However, existing computational patient stratification methods have been benchmarked almost ex
Externí odkaz:
http://arxiv.org/abs/2408.00200
Languages continually evolve in response to societal events, resulting in new terms and shifts in meanings. These changes have significant implications for computer applications, including automatic translation and chatbots, making it essential to ch
Externí odkaz:
http://arxiv.org/abs/2407.16624
We introduce CYBER-0, the first zero-order optimization algorithm for memory-and-communication efficient Federated Learning, resilient to Byzantine faults. We show through extensive numerical experiments on the MNIST dataset and finetuning RoBERTa-La
Externí odkaz:
http://arxiv.org/abs/2406.14362
Post-training quantization (PTQ) reduces the memory footprint of LLMs by quantizing weights to low-precision datatypes. Since LLM inference is usually memory-bound, PTQ methods can improve inference throughput. Recent state-of-the-art PTQ approaches
Externí odkaz:
http://arxiv.org/abs/2406.11235
Autor:
Aamir, M., Acar, B., Adamov, G., Adams, T., Adloff, C., Afanasiev, S., Agrawal, C., Ahmad, A., Ahmed, H. A., Akbar, S., Akchurin, N., Akgul, B., Akgun, B., Akpinar, R. O., Aktas, E., AlKadhim, A., Alexakhin, V., Alimena, J., Alison, J., Alpana, A., Alshehri, W., Dominguez, P. Alvarez, Alyari, M., Amendola, C., Amir, R. B., Andersen, S. B., Andreev, Y., Antoszczuk, P. D., Aras, U., Ardila, L., Aspell, P., Avila, M., Awad, I., Aydilek, O., Azimi, Z., Pretel, A. Aznar, Bach, O. A., Bainbridge, R., Bakshi, A., Bam, B., Banerjee, S., Barney, D., Bayraktar, O., Beaudette, F., Beaujean, F., Becheva, E., Behera, P. K., Belloni, A., Bergauer, T., Besancon, M., Bylund, O. Bessidskaia, Bhatt, L., Bhowmil, D., Blekman, F., Blinov, P., Bloch, P., Bodek, A., Boger, a., Bonnemaison, A., Bouyjou, F., Brennan, L., Brondolin, E., Brusamolino, A., Bubanja, I., Perraguin, A. Buchot, Bunin, P., Misura, A. Burazin, Butler-nalin, A., Cakir, A., Callier, S., Campbell, S., Canderan, K., Cankocak, K., Cappati, A., Caregari, S., Carron, S., Carty, C., Cauchois, A., Ceard, L., Cerci, S., Chang, P. J., Chatterjee, R. M., Chatterjee, S., Chattopadhyay, P., Chatzistavrou, T., Chaudhary, M. S., Chauhan, A., Chen, J. A., Chen, J., Chen, Y., Cheng, K., Cheung, H., Chhikara, J., Chiron, A., Chiusi, M., Chokheli, D., Chudasama, R., Clement, E., Mendez, S. Coco, Coko, D., Coskun, K., Couderc, F., Crossman, B., Cui, Z., Cuisset, T., Cummings, G., Curtis, E. M., D'Alfonso, M., D-hler-ball, J., Dadazhanova, O., Damgov, J., Das, I., DasGupta, S., Dauncey, P., Mendes, A. David Tinoco, Davies, G., Davignon, O., DeLa, P. deBarbaroC., DeSilva, M., DeWit, A., Debbins, P., Defranchis, M. M., Delagnes, E., Devouge, P., Dewangan, C., DiGuglielmo, G., Diehl, L., Dilsiz, K., Dincer, G. G., Dittmann, J., Dragicevic, M., Du, D., Dubinchik, B., Dugad, S., Dulucq, F., Dumanoglu, I., Duran, B., Dutta, S., Dutta, V., Dychkant, A., Dünser, M., Edberg, T., Ehle, I. T., Berni, A. El, Elias, F., Eno, S. C., Erdogan, E. N., Erkmen, B., Ershov, Y., Ertorer, E. Y., Extier, S., Eychenne, L., Fedar, Y. E., Fedi, G., De Almeida, J. P. Figueiredo De De Sá Sousa, Alves, B. A. Fontana Santos Santos, Frahm, E., Francis, K., Freeman, J., French, T., Gaede, F., Gandhi, P. K., Ganjour, S., Garcia-Bellido, A., Gastaldi, F., Gazi, L., Gecse, Z., Gerwig, H., Gevin, O., Ghosh, S., Gill, K., Gleyzer, S., Godinovic, N., Goek, M., Goettlicher, P., Goff, R., Golunov, A., Gonultas, B., Martínez, J. D. González, Gorbounov, N., Gouskos, L., Gray, A., Gray, L., Grieco, C., Groenroos, S., Groner, D., Gruber, A., Grummer, A., Grönroos, S., Guilloux, F., Guler, Y., Gungordu, A. D., Guo, J., Guo, K., Guler, E. Gurpinar, Gutti, H. K., Guvenli, A. A., Gülmez, E., Hacisahinoglu, B., Halkin, Y., Machado, G. Hamilton Ilha, Hare, H. S., Hatakeyama, K., Heering, A. H., Hegde, V., Heintz, U., Hinton, N., Hinzmann, A., Hirschauer, J., Hitlin, D., Hos, İ., Hou, B., Hou, X., Howard, A., Howe, C., Hsieh, H., Hsu, T., Hua, H., Hummer, F., Imran, M., Incandela, J., Iren, E., Isildak, B., Jackson, P. S., Jackson, W. J., Jain, S., Jana, P., Jaroslavceva, J., Jena, S., Jige, A., Jordano, P. P., Joshi, U., Kaadze, K., Kafizov, A., Kalipoliti, L., Tharayil, A. Kallil, Kaluzinska, O., Kamble, S., Kaminskiy, A., Kanemura, M., Kanso, H., Kao, Y., Kapic, A., Kapsiak, C., Karjavine, V., Karmakar, S., Karneyeu, A., Kaya, M., Topaksu, A. Kayis, Kaynak, B., Kazhykarim, Y., Khan, F. A., Khudiakov, A., Kieseler, J., Kim, R. S., Klijnsma, T., Kloiber, E. G., Klute, M., Kocak, Z., Kodali, K. R., Koetz, K., Kolberg, T., Kolcu, O. B., Komaragiri, J. R., Komm, M., Kopsalis, I., Krause, H. A., Krawczyk, M. A., Vinayakam, T. R. Krishnaswamy, Kristiansen, K., Kristic, A., Krohn, M., Kronheim, B., Krüger, K., Kudtarkar, C., Kulis, S., Kumar, M., Kumar, N., Kumar, S., Verma, R. Kumar, Kunori, S., Kunts, A., Kuo, C., Kurenkov, A., Kuryatkov, V., Kyre, S., Ladenson, J., Lamichhane, K., Landsberg, G., Langford, J., Laudrain, A., Laughlin, R., Lawhorn, J., Dortz, O. Le, Lee, S. W., Lektauers, A., Lelas, D., Leon, M., Levchuk, L., Li, A. J., Li, J., Li, Y., Liang, Z., Liao, H., Lin, K., Lin, W., Lin, Z., Lincoln, D., Linssen, L., Litomin, A., Liu, G., Liu, Y., Lobanov, A., Lohezic, V., Loiseau, T., Lu, C., Lu, R., Lu, S. Y., Lukens, P., Mackenzie, M., Magnan, A., Magniette, F., Mahjoub, A., Mahon, D., Majumder, G., Makarenko, V., Malakhov, A., Malgeri, L., Mallios, S., Mandloi, C., Mankel, A., Mannelli, M., Mans, J., Mantilla, C., Martinez, G., Massa, C., Masterson, P., Matthewman, M., Matveev, V., Mayekar, S., Mazlov, I., Mehta, A., Mestvirishvili, A., Miao, Y., Milella, G., Mirza, I. R., Mitra, P., Moccia, S., Mohanty, G. B., Monti, F., Moortgat, F., Murthy, S., Music, J., Musienko, Y., Nabili, S., Nayak, S., Nelson, J. W., Nema, A., Neutelings, I., Niedziela, J., Nikitenko, A., Noonan, D., Noy, M., Nurdan, K., Obraztsov, S., Ochando, C., Ogul, H., Olsson, J., Onel, Y., Ozkorucuklu, S., Paganis, E., Palit, P., Pan, R., Pandey, S., Pantaleo, F., Papageorgakis, C., Paramesvaran, S., Paranjpe, M. M., Parolia, S., Parsons, A. G., Parygin, P., Paulini, M., Paus, C., Peñaló, K., Pedro, K., Pekic, V., Peltola, T., Peng, B., Perego, A., Perini, D., Petrilli, A., Pham, H., Pierre-Emile, T., Podem, S. K., Popov, V., Portales, L., Potok, O., Pradeep, P. B., Pramanik, R., Prosper, H., Prvan, M., Qasim, S. R., Qu, H., Quast, T., Trivino, A. Quiroga, Rabour, L., Raicevic, N., Rajpoot, H., Rao, M. A., Rapacz, K., Redjeb, W., Reinecke, M., Revering, M., Roberts, A., Rohlf, J., Rosado, P., Rose, A., Rothman, S., Rout, P. K., Rovere, M., Rumerio, P., Rusack, R., Rygaard, L., Ryjov, V., Sadivnycha, S., Sahin, M. Ö., Sakarya, U., Salerno, R., Saradhy, R., Saraf, M., Sarbandi, K., Sarkisla, M. A., Satyshev, I., Saud, N., Sauvan, J., Schindler, G., Schmidt, A., Schmidt, I., Schmitt, M. H., Sculac, A., Sculac, T., Sedelnikov, A., Seez, C., Sefkow, F., Selivanova, D., Selvaggi, M., Sergeychik, V., Sert, H., Shahid, M., Sharma, P., Sharma, R., Sharma, S., Shelake, M., Shenai, A., Shih, C. W., Shinde, R., Shmygol, D., Shukla, R., Sicking, E., Silva, P., Simsek, C., Simsek, E., Sirasva, B. K., Sirois, Y., Song, S., Song, Y., Soudais, G., Sriram, S., StJacques, R. R., StahlLeiton, A. G., Steen, A., Stein, J., Strait, J., Strobbe, N., Su, X., Sukhov, E., Suleiman, A., Cerci, D. Sunar, Suryadevara, P., Swain, K., Syal, C., Tali, B., Tanay, K., Tang, W., Tanvir, A., Tao, J., Tarabini, A., Tatli, T., Taylor, R., Taysi, Z. C., Teafoe, G., Tee, C. Z., Terrill, W., Thienpont, D., Thomas, R., Titov, M., Todd, C., Todd, E., Toms, M., Tosun, A., Troska, J., Tsai, L., Tsamalaidze, Z., Tsionou, D., Tsipolitis, G., Tsirigoti, M., Tu, R., Polat, S. N. Tural, Undleeb, S., Usai, E., Uslan, E., Ustinov, V., Vernazza, E., Viahin, O., Viazlo, O., Vichoudis, P., Vijay, A., Virdee, T., Voirin, E., Vojinovic, M., Voytishin, N., Vámi, T. Á., Wade, A., Walter, D., Wang, C., Wang, F., Wang, J., Wang, K., Wang, X., Wang, Y., Wang, Z., Wanlin, E., Wayne, M., Wetzel, J., Whitbeck, A., Wickwire, R., Wilmot, D., Wilson, J., Wu, H., Xiao, M., Yang, J., Yazici, B., Ye, Y., Yetkin, T., Yi, R., Yohay, R., Yu, T., Yuan, C., Yuan, X., Yuksel, O., YushmanoV, I., Yusuff, I., Zabi, A., Zareckis, D., Zarubin, A., Zehetner, P., Zghiche, A., Zhang, C., Zhang, D., Zhang, H., Zhang, J., Zhang, Z., Zhao, X., Zhong, J., Zhou, Y., Zorbilmez, Ç.
A novel method to reconstruct the energy of hadronic showers in the CMS High Granularity Calorimeter (HGCAL) is presented. The HGCAL is a sampling calorimeter with very fine transverse and longitudinal granularity. The active media are silicon sensor
Externí odkaz:
http://arxiv.org/abs/2406.11937
In this paper we develop an intrinsic formalism to study the topology, smooth structure, and Riemannian geometry of the Wasserstein space of a closed Riemannian manifold. Our formalism allows for a new characterisation of the Weak topology via conver
Externí odkaz:
http://arxiv.org/abs/2406.05268
We study gradient descent (GD) dynamics on logistic regression problems with large, constant step sizes. For linearly-separable data, it is known that GD converges to the minimizer with arbitrarily large step sizes, a property which no longer holds w
Externí odkaz:
http://arxiv.org/abs/2406.05033
Autor:
Guo, Wentao, Long, Jikai, Zeng, Yimeng, Liu, Zirui, Yang, Xinyu, Ran, Yide, Gardner, Jacob R., Bastani, Osbert, De Sa, Christopher, Yu, Xiaodong, Chen, Beidi, Xu, Zhaozhuo
Zeroth-order optimization (ZO) is a memory-efficient strategy for fine-tuning Large Language Models using only forward passes. However, the application of ZO fine-tuning in memory-constrained settings such as mobile phones and laptops is still challe
Externí odkaz:
http://arxiv.org/abs/2406.02913