Abstract: Training deep learning models on medical datasets that perform well for all classes is a challenging task. It is often the case that a suboptimal performance is obtained on some classes due to the natural class imbalance issue that comes with medical data. An effective way to tackle this problem is by using targeted active learning, where we iteratively add data points to the training data that belong to the rare classes. However, existing active learning methods are ineffective in targeting rare classes in medical datasets. In this work, we propose Clinical (targeted aCtive Learning for ImbalaNced medICal imAge cLassification) a framework that uses submodular mutual information functions as acquisition functions to mine critical data points from rare classes.

Modern decision-making systems, from robots to web recommendation engines, are expected to adapt: to user preferences, changing circumstances or even new tasks. Yet, it is still uncommon to deploy a dynamically learning agent (rather than a fixed policy) to a production system, as it's perceived as unsafe. Using historical data to reason about learning algorithms, similar to offline policy evaluation (OPE) applied to fixed policies, could help practitioners evaluate and ultimately deploy such adaptive agents to production. In this work, we formalize offline learner simulation (OLS) for reinforcement learning (RL) and propose a novel evaluation protocol that measures both fidelity and efficiency of the simulation. For environments with complex high-dimensional observations, we propose a semi-parametric approach that leverages recent advances in latent state discovery in order to achieve accurate and efficient offline simulations. In preliminary experiments, we show the advantage of our approach compared to fully non-parametric baselines.

The availability of large pre-trained models is changing the landscape of Machine Learning research and practice, moving from a training-from-scratch to a fine-tuning paradigm. While in some applications the goal is to "nudge" the pre-trained distribution towards preferred outputs, in others it is to steer it towards a different distribution over the sample space. Two main paradigms have emerged to tackle this challenge: Reward Maximization (RM) and, more recently, Distribution Matching (DM). RM applies standard Reinforcement Learning (RL) techniques, such as Policy Gradients, to gradually increase the reward signal. DM prescribes to first make explicit the target distribution that the model is fine-tuned to approximate. Here we explore the theoretical connections between the two paradigms, and show that methods such as KL-control developed for RM can also be construed as belonging to DM. We further observe that while DM differs from RM, it can suffer from similar training difficulties, such as high gradient variance. We leverage connections between the two paradigms to import the concept of baseline into DM methods. We empirically validate the benefits of adding a baseline on an array of controllable language generation tasks such as constraining topic, sentiment, and gender distributions in texts sampled from a language model. We observe superior performance in terms of constraint satisfaction, stability and sample efficiency.

This month the AI Precision Health Institute at the University of Hawaiʻi Cancer Center launched a new seminar series on applications of AI in cancer research and clinical practice. The first lecture in the series was presentated by Bardia Kohn PhD. on November 4, 2022. Dr. Kohn is Associate Professor in the Mechanical Engineering Department of the University of Hawaii at Manoa and Director of the Advanced Materials and Medical Instrument Laboratory (AMMI Lab). The focus of Dr. Kohn's group is to develop new medical robotic systems to make surgeries less invasive and more accurate. Dr. Kohn presented his research on the Application of Robotics and AI in Prostate Cancer Diagnostic and Treatment Methods.

When self learning ML or DL, I have found there are a tonne of amazing courses. However, many inevitably get bogged down in the math, and the equations, and other gibberish. Not all of us intend to do research, some of us just want to have fun, and build some badass projects along the way. So, here are some courses available on the internet that teach you the pure code you need to get started with deep learning, and hopefully build some projects along the way. They are also useful if you have gained great theoretical knowledge, and would like to supplement it with great practice.

Indian Institute of Information Technology, Kottayam, which has a status equal to a university, has called for admission to MTech and PhD programmes. The MTech programmes are for employed professionals. The courses are in three much-sought-after areas and have good prospects. A pass in BTech in any of the disciplines with 50% marks, BE, AMIE or MCA or CS/IT/Maths. The candidate should be employed in an industrial establishment or in the academic sector. The maximum number of seats is 60 with a minimum of 20 seats in each branch of study.

Are you looking for a step-by-step Data Analysis with Python Roadmap? If yes, then this article is for you. In this article, you will find a step-by-step roadmap to learn data analysis with python. Along with that, at each step, you will find resources to learn. So without any further ado, let's get started- So, you have chosen Python programming for Data Analysis.

We investigate the Multi-Armed Bandit problem with Temporally-Partitioned Rewards (TP-MAB) setting in this paper. In the TP-MAB setting, an agent will receive subsets of the reward over multiple rounds rather than the entire reward for the arm all at once. In this paper, we introduce a general formulation of how an arm's cumulative reward is distributed across several rounds, called Beta-spread property. Such a generalization is needed to be able to handle partitioned rewards in which the maximum reward per round is not distributed uniformly across rounds. We derive a lower bound on the TP-MAB problem under the assumption that Beta-spread holds. Moreover, we provide an algorithm TP-UCB-FR-G, which uses the Beta-spread property to improve the regret upper bound in some scenarios. By generalizing how the cumulative reward is distributed, this setting is applicable in a broader range of applications.

Online continual learning (OCL) aims to train neural networks incrementally from a non-stationary data stream with a single pass through data. Rehearsal-based methods attempt to approximate the observed input distributions over time with a small memory and revisit them later to avoid forgetting. Despite their strong empirical performance, rehearsal methods still suffer from a poor approximation of past data's loss landscape with memory samples. This paper revisits the rehearsal dynamics in online settings. We provide theoretical insights on the inherent memory overfitting risk from the viewpoint of biased and dynamic empirical risk minimization, and examine the merits and limits of repeated rehearsal. Inspired by our analysis, a simple and intuitive baseline, repeated augmented rehearsal (RAR), is designed to address the underfitting-overfitting dilemma of online rehearsal. Surprisingly, across four rather different OCL benchmarks, this simple baseline outperforms vanilla rehearsal by 9%-17% and also significantly improves the state-of-the-art rehearsal-based methods MIR, ASER, and SCR. We also demonstrate that RAR successfully achieves an accurate approximation of the loss landscape of past data and high-loss ridge aversion in its learning trajectory. Extensive ablation studies are conducted to study the interplay between repeated and augmented rehearsal, and reinforcement learning (RL) is applied to dynamically adjust the hyperparameters of RAR to balance the stability-plasticity trade-off online.

In September, UNESCO released a report that highlighted the need for India to leverage Artificial Intelligence (AI). The stark digital divide in India however has prompted many experts to express doubts if AI can really solve the problems of access and equity. While there may be some truth in the apprehensions, perhaps a deeper issue is whether the EdTech sector has even tried to leverage AI to address the needs of India beyond the "upper middle class". Post pandemic India is ready for innovations that can address this gap and drive the EdTech industry forward. Since its emergence in the mid-1950s, AI has grown considerably. Along with big data, blockchain, augmented reality and the Internet of Things (IoT), AI has been hailed for ushering in the Fourth Industrial Revolution or Industry 4.0.