More Like this

1. Mode-Selective Frequency Conversion in a Three-Mode Fiber

   https://doi.org/10.1364/CLEO_SI.2020.SM3P.3  

   Shamsshooli, Afshin; Guo, Cheng; Parmigiani, Francesca; Li, Xiaoying; Vasilyev, Michael (January 2020, CLEO 2020 Conference Proceedings)

   We present a scheme for spatial-mode-selective frequency conversion in a few-mode fiber and experimentally demonstrate upconversion of either of two signal modes from C-band to fundamental mode in S-band with crosstalk below –15.5 dB. 

   more » « less
2. Reconfigurable mode-selective frequency conversion in a three-mode fiber

   https://doi.org/10.1109/IPC47351.2020.9252379  

   Shamsshooli, Afshin; Guo, Cheng; Vasilyev, Michael; Parmigiani, Francesca; Li, Xiaoying (October 2020, IEEE Photonics Conference 2020)

   null (Ed.)

   We present a scheme for spatial-mode-selective frequency conversion in a few-mode fiber and experimentally demonstrate upconversion of arbitrary superpositions of two signal modes from C-band to the fundamental mode in S-band with conversion efficiencies within 1 dB range of one another. 

   more » « less
3. B-mode Targets

   Kallosh, Renata; Linde, Andrei (July 2019, ArXiv.org)
4. Mechanistic Mode Connectivity

   Lubana, Ekdeep Singh; Eric J. Bigelow; Robert P. Dick; David Krueger; Hidenori Tanaka. (July 2023, Proc. Int. Conf. on Machine Learning)

   We study neural network loss landscapes through the lens of mode connectivity, the observation that minimizers of neural networks retrieved via training on a dataset are connected via simple paths of low loss. Specifically, we ask the following question: are minimizers that rely on different mechanisms for making their predictions connected via simple paths of low loss? We provide a definition of mechanistic similarity as shared invariances to input transformations and demonstrate that lack of linear connectivity between two models implies they use dissimilar mechanisms for making their predictions. Relevant to practice, this result helps us demonstrate that naive fine-tuning on a downstream dataset can fail to alter a model’s mechanisms, e.g., fine-tuning can fail to eliminate a model’s reliance on spurious attributes. Our analysis also motivates a method for targeted alteration of a model’s mechanisms, named connectivity-based fine-tuning (CBFT), which we analyze using several synthetic datasets for the task of reducing a model’s reliance on spurious attributes. 

   more » « less
5. Antarctic Mode Water

   Speer, K. (January 2018, Journal of Marine Research)

   null (Ed.)