-
QGFN: Controllable Greediness with Action Values
Authors:
Elaine Lau,
Stephen Zhewen Lu,
Ling Pan,
Doina Precup,
Emmanuel Bengio
Abstract:
Generative Flow Networks (GFlowNets; GFNs) are a family of reward/energy-based generative methods for combinatorial objects, capable of generating diverse and high-utility samples. However, biasing GFNs towards producing high-utility samples is non-trivial. In this work, we leverage connections between GFNs and reinforcement learning (RL) and propose to combine the GFN policy with an action-value…
▽ More
Generative Flow Networks (GFlowNets; GFNs) are a family of reward/energy-based generative methods for combinatorial objects, capable of generating diverse and high-utility samples. However, biasing GFNs towards producing high-utility samples is non-trivial. In this work, we leverage connections between GFNs and reinforcement learning (RL) and propose to combine the GFN policy with an action-value estimate, $Q$, to create greedier sampling policies which can be controlled by a mixing parameter. We show that several variants of the proposed method, QGFN, are able to improve on the number of high-reward samples generated in a variety of tasks without sacrificing diversity.
△ Less
Submitted 23 May, 2024; v1 submitted 7 February, 2024;
originally announced February 2024.
-
Screening of Pneumonia and Urinary Tract Infection at Triage using TriNet
Authors:
Stephen Z. Lu
Abstract:
Due to the steady rise in population demographics and longevity, emergency department visits are increasing across North America. As more patients visit the emergency department, traditional clinical workflows become overloaded and inefficient, leading to prolonged wait-times and reduced healthcare quality. One of such workflows is the triage medical directive, impeded by limited human workload, i…
▽ More
Due to the steady rise in population demographics and longevity, emergency department visits are increasing across North America. As more patients visit the emergency department, traditional clinical workflows become overloaded and inefficient, leading to prolonged wait-times and reduced healthcare quality. One of such workflows is the triage medical directive, impeded by limited human workload, inaccurate diagnoses and invasive over-testing. To address this issue, we propose TriNet: a machine learning model for medical directives that automates first-line screening at triage for conditions requiring downstream testing for diagnosis confirmation. To verify screening potential, TriNet was trained on hospital triage data and achieved high positive predictive values in detecting pneumonia (0.86) and urinary tract infection (0.93). These models outperform current clinical benchmarks, indicating that machine-learning medical directives can offer cost-free, non-invasive screening with high specificity for common conditions, reducing the risk of over-testing while increasing emergency department efficiency.
△ Less
Submitted 5 September, 2023;
originally announced September 2023.
-
Pushing the feature size down to 11 nm by interference photolithography with hyperbolic metamaterials under conventional UV light source
Authors:
Xuefeng Yang,
Shuxia Zhang,
Baoji Wang,
Xiaolin Cai,
Xiaohua Li,
Weiyang Yu,
Qin Wang,
snd Zhongliang Lu
Abstract:
Limited by the cost and complexity, superresolution lithography is hard to achieve through the traditional interference lithography. We here developed the plasmonic interference lithography technique by using a hyperbolic metamaterials (HMMs) / photoresist / metal plasmonic waveguide to push the feature sizes theoretically down to 16 nm and even to 11 nm at the wavelength of 365 nm with TM polariz…
▽ More
Limited by the cost and complexity, superresolution lithography is hard to achieve through the traditional interference lithography. We here developed the plasmonic interference lithography technique by using a hyperbolic metamaterials (HMMs) / photoresist / metal plasmonic waveguide to push the feature sizes theoretically down to 16 nm and even to 11 nm at the wavelength of 365 nm with TM polarization. The waveguide based on the proposed HMMs can support high-k mode for superresolution lithography. Furthermore, plasmonic mode supported in the proposed lithography structure can be tailored by dimension of HMM and permittivity of the materials, which makes it possible to get higher resolution pattern under conventional UV light. Our findings open up an avenue to pushing the nanolithography node towards 10 nm for low-cost and large area fabrication under conventional UV light source.
△ Less
Submitted 6 September, 2022;
originally announced September 2022.
-
Weighted multilinear Hardy operators and commutators
Authors:
Zun Wei Fu,
Shu Li Gong,
Shan Zhen Lu,
Wen Yuan
Abstract:
In this paper, we introduce a type of weighted multilinear Hardy operators and obtain their sharp bounds on the product of Lebesgue spaces and central Morrey spaces. In addition, we obtain sufficient and necessary conditions of the weight functions so that the commutators of the weighted multilinear Hardy operators (with symbols in central BMO space) are bounded on the product of central Morrey sp…
▽ More
In this paper, we introduce a type of weighted multilinear Hardy operators and obtain their sharp bounds on the product of Lebesgue spaces and central Morrey spaces. In addition, we obtain sufficient and necessary conditions of the weight functions so that the commutators of the weighted multilinear Hardy operators (with symbols in central BMO space) are bounded on the product of central Morrey spaces. These results are further used to prove sharp estimates of some inequalities due to Riemann-Liouville and Weyl.
△ Less
Submitted 20 September, 2014;
originally announced September 2014.
