Centuries are a celebrated event in cricket, usually resulting in match-winning innings by the batsman. As a statistics enthusiast, it felt like a great problem to model because it is not only immensely interesting, the novelty of the problem did make it challenging. This piece explains the reasoning behind how I prepared the data, what model I used, and the evaluation criteria. In a previous post, I did a probabilistic analysis of centuries, a key finding was that unconditioned on anything else, the empirically estimated probability of a batsman knock resulting in a century is only 3.16%. This is important because when modeling a classification problem, class prevalence is probably the most crucial factor in determining the efficacy of your model(s).

Despite being high dimensional, physics datasets are highly structured since physical laws strictly govern the data generating process. Although the data is complicated, it is not hard to imagine that physics data can exist within low-dimensional manifolds inside a high-dimensional ambient space. There is a growing recent interest in endowing the space of collider events with a metric structure calculated directly in the space of its inputs. Metrics based on optimal transport, such as energy mover's distance (EMD) [1] and Hellinger distance [2], allow us to compare raw inputs directly and quantify the global structural difference between any pair of collider events. Since the advent of these studies, a broad range of use cases has been emerging for these metrics. These include event tagging, anomaly tagging[3-5], and measurements of Quantum Chromo Dynamical (QCD) properties. However, the input dimension is usually very large for collider data; thus, the induced manifold of the metric lives in a very high dimensional space, making it challenging to work with directly.

Federated learning offers a framework of training a machine learning model in a distributed fashion while preserving privacy of the participants. As the server cannot govern the clients' actions, nefarious clients may attack the global model by sending malicious local gradients. In the meantime, there could also be unreliable clients who are benign but each has a portion of low-quality training data (e.g., blur or low-resolution images), thus may appearing similar as malicious clients. Therefore, a defense mechanism will need to perform a three-fold differentiation which is much more challenging than the conventional (two-fold) case. This paper introduces MUD-HoG, a novel defense algorithm that addresses this challenge in federated learning using long-short history of gradients, and treats the detected malicious and unreliable clients differently. Not only this, but we can also distinguish between targeted and untargeted attacks among malicious clients, unlike most prior works which only consider one type of the attacks. Specifically, we take into account sign-flipping, additive-noise, label-flipping, and multi-label-flipping attacks, under a non-IID setting. We evaluate MUD-HoG with six state-of-the-art methods on two datasets. The results show that MUD-HoG outperforms all of them in terms of accuracy as well as precision and recall, in the presence of a mixture of multiple (four) types of attackers as well as unreliable clients. Moreover, unlike most prior works which can only tolerate a low population of harmful users, MUD-HoG can work with and successfully detect a wide range of malicious and unreliable clients - up to 47.5% and 10%, respectively, of the total population. Our code is open-sourced at https://github.com/LabSAINT/MUD-HoG_Federated_Learning.

Yoga is a globally acclaimed and widely recommended practice for a healthy living. Maintaining correct posture while performing a Yogasana is of utmost importance. In this work, we employ transfer learning from Human Pose Estimation models for extracting 136 key-points spread all over the body to train a Random Forest classifier which is used for estimation of the Yogasanas. The results are evaluated on an in-house collected extensive yoga video database of 51 subjects recorded from 4 different camera angles. We propose a 3 step scheme for evaluating the generalizability of a Yoga classifier by testing it on 1) unseen frames, 2) unseen subjects, and 3) unseen camera angles. We argue that for most of the applications, validation accuracies on unseen subjects and unseen camera angles would be most important. We empirically analyze over three public datasets, the advantage of transfer learning and the possibilities of target leakage. We further demonstrate that the classification accuracies critically depend on the cross validation method employed and can often be misleading. To promote further research, we have made key-points dataset and code publicly available.

The model performance is based on dividing the known data into two parts, one to train the model and the other to test the prediction performance, thus obtaining the model accuracy and adjusting it according to the results. However, accuracy depends on how we slip the data, which can lead to possible biases in the model that prevent accuracy from generalizing to unseen data. Cross-validation is used to combat the random split of the data. This is a method that allows testing the performance of a predictive machine learning model, based on the same principle of the Train-Test split technique but with the difference that it must be performed k times and obtain the accuracy of each attempt. This technique is known as k-folds, where each fold is a specific division of the data different from the rest.

Purpose; The purpose of this study is to classify glial tumors into grade II, III and IV categories noninvasively by application of machine learning to multi-modal MRI features in comparison with volumetric analysis. Methods; We retrospectively studied 57 glioma patients with pre and postcontrast T1 weighted, T2 weighted, FLAIR images, and ADC maps acquired on a 3T MRI. The tumors were segmented into enhancing and nonenhancing portions, tumor necrosis, cyst and edema using semiautomated segmentation of ITK-SNAP open source tool. We measured total tumor volume, enhancing-nonenhancing tumor, edema, necrosis volume and the ratios to the total tumor volume. Training of a support vector machine (SVM) classifier and artificial neural network (ANN) was performed with labeled data designed to answer the question of interest. Specificity, sensitivity, and AUC of the predictions were computed by means of ROC analysis. Differences in continuous measures between groups were assessed by using Kruskall Wallis, with post hoc Dunn correction for multiple comparisons. Results; When we compared the volume ratios between groups, there was statistically significant difference between grade IV and grade II-III glial tumors. Edema and tumor necrosis volume ratios for grade IV glial tumors were higher than that of grade II and III. Volumetric ratio analysis could not distinguish grade II and III tumors successfully. However, SVM and ANN correctly classified each group with accuracies up to 98% and 96%. Conclusion; Application of machine learning methods to MRI features can be used to classify brain tumors noninvasively and more readily in clinical settings.

We introduce a new type of format-transforming encryption where the format of ciphertexts is implicitly encoded within a machine-learned generative model. Around this primitive, we build a system for covert messaging over large, public internet platforms (e.g., Twitter). Loosely, our system composes an authenticated encryption scheme, with a method for encoding random ciphertext bits into samples from the generative model's family of seed-indexed token-distributions. By fixing a deployment scenario, we are forced to consider system-level and algorithmic solutions to real challenges -- ~such as receiver-side parsing ambiguities, and the low information-carrying capacity of actual token-distributions~ -- that were elided in prior work. We use GPT-2 as our generative model so that our system cryptographically transforms plaintext bitstrings into natural-language covertexts suitable for posting to public platforms. We consider adversaries with full view of the internet platform's content, whose goal is to surface posts that are using our system for covert messaging. We carry out a suite of experiments to provide heuristic evidence of security and to explore tradeoffs between operational efficiency and detectability.

Unmanned Aerial Vehicle (UAV) has gained significant traction in the recent years, particularly the context of surveillance. However, video datasets that capture violent and non-violent human activity from aerial point-of-view is scarce. To address this issue, we propose a novel, baseline simulator which is capable of generating sequences of photo-realistic synthetic images of crowds engaging in various activities that can be categorized as violent or non-violent. The crowd groups are annotated with bounding boxes that are automatically computed using semantic segmentation. Our simulator is capable of generating large, randomized urban environments and is able to maintain an average of 25 frames per second on a mid-range computer with 150 concurrent crowd agents interacting with each other. We also show that when synthetic data from the proposed simulator is augmented with real world data, binary video classification accuracy is improved by 5% on average across two different models.

Malicious software (malware) classification offers a unique challenge for continual learning (CL) regimes due to the volume of new samples received on a daily basis and the evolution of malware to exploit new vulnerabilities. On a typical day, antivirus vendors receive hundreds of thousands of unique pieces of software, both malicious and benign, and over the course of the lifetime of a malware classifier, more than a billion samples can easily accumulate. Given the scale of the problem, sequential training using continual learning techniques could provide substantial benefits in reducing training and storage overhead. To date, however, there has been no exploration of CL applied to malware classification tasks. In this paper, we study 11 CL techniques applied to three malware tasks covering common incremental learning scenarios, including task, class, and domain incremental learning (IL). Specifically, using two realistic, large-scale malware datasets, we evaluate the performance of the CL methods on both binary malware classification (Domain-IL) and multi-class malware family classification (Task-IL and Class-IL) tasks. To our surprise, continual learning methods significantly underperformed naive Joint replay of the training data in nearly all settings -- in some cases reducing accuracy by more than 70 percentage points. A simple approach of selectively replaying 20% of the stored data achieves better performance, with 50% of the training time compared to Joint replay. Finally, we discuss potential reasons for the unexpectedly poor performance of the CL techniques, with the hope that it spurs further research on developing techniques that are more effective in the malware classification domain.

Datacenters execute large computational jobs, which are composed of smaller tasks. A job completes when all its tasks finish, so stragglers -- rare, yet extremely slow tasks -- are a major impediment to datacenter performance. Accurately predicting stragglers would enable proactive intervention, allowing datacenter operators to mitigate stragglers before they delay a job. While much prior work applies machine learning to predict computer system performance, these approaches rely on complete labels -- i.e., sufficient examples of all possible behaviors, including straggling and non-straggling -- or strong assumptions about the underlying latency distributions -- e.g., whether Gaussian or not. Within a running job, however, none of this information is available until stragglers have revealed themselves when they have already delayed the job. To predict stragglers accurately and early without labeled positive examples or assumptions on latency distributions, this paper presents NURD, a novel Negative-Unlabeled learning approach with Reweighting and Distribution-compensation that only trains on negative and unlabeled streaming data. The key idea is to train a predictor using finished tasks of non-stragglers to predict latency for unlabeled running tasks, and then reweight each unlabeled task's prediction based on a weighting function of its feature space. We evaluate NURD on two production traces from Google and Alibaba, and find that compared to the best baseline approach, NURD produces 2--11 percentage point increases in the F1 score in terms of prediction accuracy, and 2.0--8.8 percentage point improvements in job completion time.