Devastating truth about Rob Reiner's daughter Romy: Her own addiction battle... how she'lived in fear' of Nick... and the handsome companion she's leaning on, all revealed by heartbroken friends Baby-faced accused killers will be tried as adults after 14-year-old girl's horrific murder Trans killer, 30, who executed her parents then converted to Islam is jailed for 25 years after trying to skip'stressful' sentencing I'm Miley Cyrus's REAL mother: Woman at center of bombshell'adoption' lawsuit breaks silence about'pregnancy at age 12' and makes MORE wild claims School bus driver responds to backlash after she was fired over'English-only' sign Six common medications you should NEVER mix with alcohol: Doctors reveal how that'pre-emptive' painkiller could destroy your liver... and the most deadly combination of all Next domino falls in Michigan's Sherrone Moore scandal as top assistant defects to SEC school The extravagant gifts the rich are buying this Christmas including an'extra person' in their marriage I was forced into Witness Protection at age seven... here's how the program nearly ruined my life Former Nickelodeon star is now'homeless on the streets of Los Angeles' How Tom Brady REALLY feels about Gisele Bundchen's secret wedding to jiu-jitsu instructor... as insiders whisper about potential of his OWN second marriage The hidden blueprint to keep MAGA in power for 100 years as Trump's inner circle shows signs of cracking Kimberly Guilfoyle's'yelling fit' after ex Donald Trump Jr's new engagement... as insiders reveal her nasty texts and derogatory nickname for Bettina Anderson Smug because you love your job? READ MORE: Scientists reveal surprising secret behind Bill Gates' success The saying goes, if you find a job you love you'll never work a day in your life. But an expert has now warned that this can backfire - and the seemingly innocent idea of loving your work can take on a moral edge. Mijeong Kwon, assistant professor of management at Rice University in Texas, said the dream of enjoying your career has become compulsive for many. 'Working for money, prestige or family obligation starts to look less admirable, even suspect,' she wrote on The Conversation .

As the popularity of deep learning (DL) in the field of magnetic resonance imaging (MRI) continues to rise, recent research has indicated that DL-based MRI reconstruction models might be excessively sensitive to minor input disturbances, including worst-case additive perturbations. This sensitivity often leads to unstable, aliased images. This raises the question of how to devise DL techniques for MRI reconstruction that can be robust to train-test variations. To address this problem, we propose a novel image reconstruction framework, termed Smoothed Unrolling (SMUG), which advances a deep unrolling-based MRI reconstruction model using a randomized smoothing (RS)-based robust learning approach. RS, which improves the tolerance of a model against input noises, has been widely used in the design of adversarial defense approaches for image classification tasks. Yet, we find that the conventional design that applies RS to the entire DL-based MRI model is ineffective. In this paper, we show that SMUG and its variants address the above issue by customizing the RS process based on the unrolling architecture of a DL-based MRI reconstruction model. Compared to the vanilla RS approach, we show that SMUG improves the robustness of MRI reconstruction with respect to a diverse set of instability sources, including worst-case and random noise perturbations to input measurements, varying measurement sampling rates, and different numbers of unrolling steps. Furthermore, we theoretically analyze the robustness of our method in the presence of perturbations.

Although deep learning (DL) has gained much popularity for accelerated magnetic resonance imaging (MRI), recent studies have shown that DL-based MRI reconstruction models could be oversensitive to tiny input perturbations (that are called 'adversarial perturbations'), which cause unstable, low-quality reconstructed images. This raises the question of how to design robust DL methods for MRI reconstruction. To address this problem, we propose a novel image reconstruction framework, termed SMOOTHED UNROLLING (SMUG), which advances a deep unrolling-based MRI reconstruction model using a randomized smoothing (RS)-based robust learning operation. RS, which improves the tolerance of a model against input noises, has been widely used in the design of adversarial defense for image classification. Yet, we find that the conventional design that applies RS to the entire DL process is ineffective for MRI reconstruction. We show that SMUG addresses the above issue by customizing the RS operation based on the unrolling architecture of the DL-based MRI reconstruction model. Compared to the vanilla RS approach and several variants of SMUG, we show that SMUG improves the robustness of MRI reconstruction with respect to a diverse set of perturbation sources, including perturbations to the input measurements, different measurement sampling rates, and different unrolling steps. Code for SMUG will be available at https://github.com/LGM70/SMUG.

Symbolic techniques based on Satisfiability Modulo Theory (SMT) solvers have been proposed for analyzing and verifying neural network properties, but their usage has been fairly limited owing to their poor scalability with larger networks. In this work, we propose a technique for combining gradient-based methods with symbolic techniques to scale such analyses and demonstrate its application for model explanation. In particular, we apply this technique to identify minimal regions in an input that are most relevant for a neural network's prediction. Our approach uses gradient information (based on Integrated Gradients) to focus on a subset of neurons in the first layer, which allows our technique to scale to large networks. The corresponding SMT constraints encode the minimal input mask discovery problem such that after masking the input, the activations of the selected neurons are still above a threshold. After solving for the minimal masks, our approach scores the mask regions to generate a relative ordering of the features within the mask. This produces a saliency map which explains "where a model is looking" when making a prediction. We evaluate our technique on three datasets - MNIST, ImageNet, and Beer Reviews, and demonstrate both quantitatively and qualitatively that the regions generated by our approach are sparser and achieve higher saliency scores compared to the gradient-based methods alone.