Search

Computer Architecture, broadly, involves optimizing hardware and software for current and future processing systems. Although there are several other top venues to publish Computer Architecture research, including ASPLOS, HPCA, and MICRO, ISCA (the International Symposium on Computer Architecture) is one of the oldest, longest running, and most prestigious venues for publishing Computer Architecture research. Since 1973, except for 1975, ISCA has been organized annually. Accordingly, this year will be the 50th year of ISCA. Thus, we set out to analyze the past 50 years of ISCA to understand who and what has been driving and innovating computing systems thus far. Our analysis identifies several interesting trends that reflect how ISCA, and Computer Architecture in general, has grown and evolved in the past 50 years, including minicomputers, general-purpose uniprocessor CPUs, multiprocessor and multi-core CPUs, general-purpose GPUs, and accelerators., Comment: 34 pages, 16 figures

Over one in three people are affected by neurodegenerative disorders. Neural stem cells, which are multipotent regenerative cells with the potential to differentiate into any of the neural cell types, have immense therapeutic potential for treating neurological disorders. However, lengthy differentiation protocols hinder clinical applications and research. In this study, we present a deep learning approach using convolutional neural networks (CNNs) to predict the fate of neural stem cell differentiation at an early stage. We trained a CNN model on a dataset of cellular images from neural stem cell cultures. Our models achieved impressive results in predicting neuron and glial cell differentiation, with a 93.3% testing accuracy for a multiclass Resnet50 model (and 99.7% accuracy for a binary Resnet50 model). In addition, we developed and published a web tool to give stem cell researchers access to this technology to allow for efficient prediction of stem cell cell differentiation. Our work demonstrates the feasibility of and builds tooling for using CNNs for rapid, early differentiation outcome prediction from simple microscopy images, which could greatly accelerate neural stem cell research and therapies.

Self-attention (SA) mechanisms have been widely used in developing sequential recommendation (SR) methods, and demonstrated state-of-the-art performance. However, in this paper, we show that self-attentive SR methods substantially suffer from the over-smoothing issue that item embeddings within a sequence become increasingly similar across attention blocks. As widely demonstrated in the literature, this issue could lead to a loss of information in individual items, and significantly degrade models' scalability and performance. To address the over-smoothing issue, in this paper, we view items within a sequence constituting a star graph and develop a method, denoted as MSSG, for SR. Different from existing self-attentive methods, MSSG introduces an additional internal node to specifically capture the global information within the sequence, and does not require information propagation among items. This design fundamentally addresses the over-smoothing issue and enables MSSG a linear time complexity with respect to the sequence length. We compare MSSG with ten state-of-the-art baseline methods on six public benchmark datasets. Our experimental results demonstrate that MSSG significantly outperforms the baseline methods, with an improvement of as much as 10.10%. Our analysis shows the superior scalability of MSSG over the state-of-the-art self-attentive methods. Our complexity analysis and run-time performance comparison together show that MSSG is both theoretically and practically more efficient than self-attentive methods. Our analysis of the attention weights learned in SA-based methods indicates that on sparse recommendation data, modeling dependencies in all item pairs using the SA mechanism yields limited information gain, and thus, might not benefit the recommendation performance, Comment: arXiv admin note: text overlap with arXiv:2209.07997

The use of Evolutionary Algorithms (EA) for solving Mathematical/Computational Optimization Problems is inspired by the biological processes of Evolution. Few of the primitives involved in the Evolutionary process/paradigm are selection of 'Fit' individuals (from a population sample) for retention, cloning, mutation, discarding, breeding, crossover etc. In the Evolutionary Algorithm abstraction, the individuals are deemed to be solution candidates to an Optimization problem and additional solution(/sets) are built by applying analogies to the above primitives (cloning, mutation etc.) by means of evaluating a 'Fitness' function/criterion. One such algorithm is Differential Evolution (DE) which can be used to compute the minima of functions such as the rastrigin function and rosenbrock function. This work is an attempt to study the result of applying the DE method on these functions with candidate individuals generated on classical Turing modeled computation and comparing the same with those on state of the art Quantum computation.The study benchmarks the convergence of these functions by varying the parameters initialized and reports timing, convergence, and resource utilization results.

In recent years, with large language models (LLMs) achieving state-of-the-art performance in context understanding, increasing efforts have been dedicated to developing LLM-enhanced sequential recommendation (SR) methods. Considering that most existing LLMs are not specifically optimized for recommendation tasks, adapting them for SR becomes a critical step in LLM-enhanced SR methods. Though numerous adaptation methods have been developed, it still remains a significant challenge to adapt LLMs for SR both efficiently and effectively. To address this challenge, in this paper, we introduce a novel side sequential network adaptation method, denoted as SSNA, for LLM enhanced SR. SSNA features three key designs to allow both efficient and effective LLM adaptation. First, SSNA learns adapters separate from LLMs, while fixing all the pre-trained parameters within LLMs to allow efficient adaptation. In addition, SSNA adapts the top-a layers of LLMs jointly, and integrates adapters sequentially for enhanced effectiveness (i.e., recommendation performance). We compare SSNA against five state-of-the-art baseline methods on five benchmark datasets using three LLMs. The experimental results demonstrate that SSNA significantly outperforms all the baseline methods in terms of recommendation performance, and achieves substantial improvement over the best-performing baseline methods at both run-time and memory efficiency during training. Our analysis shows the effectiveness of integrating adapters in a sequential manner. Our parameter study demonstrates the effectiveness of jointly adapting the top-a layers of LLMs.

As software projects progress, quality of code assumes paramount importance as it affects reliability, maintainability and security of software. For this reason, static analysis tools are used in developer workflows to flag code quality issues. However, developers need to spend extra efforts to revise their code to improve code quality based on the tool findings. In this work, we investigate the use of (instruction-following) large language models (LLMs) to assist developers in revising code to resolve code quality issues. We present a tool, CORE (short for COde REvisions), architected using a pair of LLMs organized as a duo comprised of a proposer and a ranker. Providers of static analysis tools recommend ways to mitigate the tool warnings and developers follow them to revise their code. The \emph{proposer LLM} of CORE takes the same set of recommendations and applies them to generate candidate code revisions. The candidates which pass the static quality checks are retained. However, the LLM may introduce subtle, unintended functionality changes which may go un-detected by the static analysis. The \emph{ranker LLM} evaluates the changes made by the proposer using a rubric that closely follows the acceptance criteria that a developer would enforce. CORE uses the scores assigned by the ranker LLM to rank the candidate revisions before presenting them to the developer. CORE could revise 59.2% Python files (across 52 quality checks) so that they pass scrutiny by both a tool and a human reviewer. The ranker LLM is able to reduce false positives by 25.8% in these cases. CORE produced revisions that passed the static analysis tool in 76.8% Java files (across 10 quality checks) comparable to 78.3% of a specialized program repair tool, with significantly much less engineering efforts.

Software engineering activities such as package migration, fixing errors reports from static analysis or testing, and adding type annotations or other specifications to a codebase, involve pervasively editing the entire repository of code. We formulate these activities as repository-level coding tasks. Recent tools like GitHub Copilot, which are powered by Large Language Models (LLMs), have succeeded in offering high-quality solutions to localized coding problems. Repository-level coding tasks are more involved and cannot be solved directly using LLMs, since code within a repository is inter-dependent and the entire repository may be too large to fit into the prompt. We frame repository-level coding as a planning problem and present a task-agnostic framework, called CodePlan to solve it. CodePlan synthesizes a multi-step chain of edits (plan), where each step results in a call to an LLM on a code location with context derived from the entire repository, previous code changes and task-specific instructions. CodePlan is based on a novel combination of an incremental dependency analysis, a change may-impact analysis and an adaptive planning algorithm. We evaluate the effectiveness of CodePlan on two repository-level tasks: package migration (C#) and temporal code edits (Python). Each task is evaluated on multiple code repositories, each of which requires inter-dependent changes to many files (between 2-97 files). Coding tasks of this level of complexity have not been automated using LLMs before. Our results show that CodePlan has better match with the ground truth compared to baselines. CodePlan is able to get 5/6 repositories to pass the validity checks (e.g., to build without errors and make correct code edits) whereas the baselines (without planning but with the same type of contextual information as CodePlan) cannot get any of the repositories to pass them.

Learning effective recommendation models from sparse user interactions represents a fundamental challenge in developing sequential recommendation methods. Recently, pre-training-based methods have been developed to tackle this challenge. Though promising, in this paper, we show that existing methods suffer from the notorious negative transfer issue, where the model adapted from the pre-trained model results in worse performance compared to the model learned from scratch in the task of interest (i.e., target task). To address this issue, we develop a method, denoted as ANT, for transferable sequential recommendation. ANT mitigates negative transfer by 1) incorporating multi-modality item information, including item texts, images and prices, to effectively learn more transferable knowledge from related tasks (i.e., auxiliary tasks); and 2) better capturing task-specific knowledge in the target task using a re-learning-based adaptation strategy. We evaluate ANT against eight state-of-the-art baseline methods on five target tasks. Our experimental results demonstrate that ANT does not suffer from the negative transfer issue on any of the target tasks. The results also demonstrate that ANT substantially outperforms baseline methods in the target tasks with an improvement of as much as 15.2%. Our analysis highlights the superior effectiveness of our re-learning-based strategy compared to fine-tuning on the target tasks.

The Australian, Chinese, European, Indian, and North American pulsar timing array (PTA) collaborations recently reported, at varying levels, evidence for the presence of a nanohertz gravitational wave background (GWB). Given that each PTA made different choices in modeling their data, we perform a comparison of the GWB and individual pulsar noise parameters across the results reported from the PTAs that constitute the International Pulsar Timing Array (IPTA). We show that despite making different modeling choices, there is no significant difference in the GWB parameters that are measured by the different PTAs, agreeing within $1\sigma$. The pulsar noise parameters are also consistent between different PTAs for the majority of the pulsars included in these analyses. We bridge the differences in modeling choices by adopting a standardized noise model for all pulsars and PTAs, finding that under this model there is a reduction in the tension in the pulsar noise parameters. As part of this reanalysis, we "extended" each PTA's data set by adding extra pulsars that were not timed by that PTA. Under these extensions, we find better constraints on the GWB amplitude and a higher signal-to-noise ratio for the Hellings and Downs correlations. These extensions serve as a prelude to the benefits offered by a full combination of data across all pulsars in the IPTA, i.e., the IPTA's Data Release 3, which will involve not just adding in additional pulsars, but also including data from all three PTAs where any given pulsar is timed by more than as single PTA., Comment: 21 pages, 9 figures, submitted to ApJ

The MMGPS-L is the most sensitive pulsar survey in the Southern Hemisphere. We present a follow-up study of one of these new discoveries, PSR J1208-5936, a 28.71-ms recycled pulsar in a double neutron star system with an orbital period of Pb=0.632 days and an eccentricity of e=0.348. Through timing of almost one year of observations, we detected the relativistic advance of periastron (0.918(1) deg/yr), resulting in a total system mass of Mt=2.586(5) Mo. We also achieved low-significance constraints on the amplitude of the Einstein delay and Shapiro delay, in turn yielding constraints on the pulsar mass (Mp=1.26(+0.13/-0.25) Mo), the companion mass (Mc=1.32(+0.25/-0.13) Mo, and the inclination angle (i=57(12) degrees). This system is highly eccentric compared to other Galactic field double neutron stars with similar periods, possibly hinting at a larger-than-usual supernova kick during the formation of the second-born neutron star. The binary will merge within 7.2(2) Gyr due to the emission of gravitational waves. With the improved sensitivity of the MMGPS-L, we updated the Milky Way neutron star merger rate to be 25(+19/-9) Myr$^{-1}$ within 90% credible intervals, which is lower than previous studies based on known Galactic binaries owing to the lack of further detections despite the highly sensitive nature of the survey. This implies a local cosmic neutron star merger rate of 293(+222/-103} Gpc/yr, consistent with LIGO and Virgo O3 observations. With this, we predict the observation of 10(+8/-4) neutron star merger events during the LIGO-Virgo-KAGRA O4 run. We predict the uncertainties on the component masses and the inclination angle will be reduced to 5x10$^{-3}$ Mo and 0.4 degrees after two decades of timing, and that in at least a decade from now the detection of the shift in Pb and the sky proper motion will serve to make an independent constraint of the distance to the system.

Aims. A realistic model of magnetic linkage between a central object and its accretion disk is a prerequisite for understanding the spin history of stars and stellar remnants. To this end, we aim to provide an analytic model in agreement with magnetohydrodynamic (MHD) simulations. Methods. For the first time, we wrote a full set of stationary asymptotic expansion equations of a thin magnetic accretion disk, including the induction and energy equations. We also performed a resistive MHD simulation of an accretion disk around a star endowed with a magnetic dipole, using the publicly available code PLUTO. We compared the analytical results with the numerical solutions, and discussed the results in the context of previous solutions of the induction equation describing the star-disk magnetospheric interaction. Results. We found that the magnetic field threading the disk is suppressed by orders of magnitude inside thin disks, so the presence of the stellar magnetic field does not strongly affect the velocity field, nor the density profile inside the disk. Density and velocity fields found in the MHD simulations match the radial and vertical profiles of the analytic solution. Qualitatively, the MHD simulations result in an internal magnetic field similar to the solutions previously obtained by solving the induction equation in the disk alone. However, the magnetic field configuration is quantitatively affected by magnetic field inflation outside the disk; this is reflected in the net torque. The torque on the star is an order of magnitude larger in the magnetic than in the non-magnetic case. Spin-up of the star occurs on a timescale comparable to the accretion timescale in the MHD case, and is an order of magnitude slower in the absence of a stellar magnetic field., Comment: 17 pages, accepted in Astronomy and Astrophysics. arXiv admin note: text overlap with arXiv:1907.12592

Ligand-based drug design aims to identify novel drug candidates of similar shapes with known active molecules. In this paper, we formulated an in silico shape-conditioned molecule generation problem to generate 3D molecule structures conditioned on the shape of a given molecule. To address this problem, we developed a translation- and rotation-equivariant shape-guided generative model ShapeMol. ShapeMol consists of an equivariant shape encoder that maps molecular surface shapes into latent embeddings, and an equivariant diffusion model that generates 3D molecules based on these embeddings. Experimental results show that ShapeMol can generate novel, diverse, drug-like molecules that retain 3D molecular shapes similar to the given shape condition. These results demonstrate the potential of ShapeMol in designing drug candidates of desired 3D shapes binding to protein target pockets.

We revisit the online portfolio allocation problem and propose universal portfolios that use factor weighing to produce portfolios that out-perform uniform dirichlet allocation schemes. We show a few analytical results on the lower bounds of portfolio growth when the returns are known to follow a factor model. We also show analytically that factor weighted dirichlet sampled portfolios dominate the wealth generated by uniformly sampled dirichlet portfolios. We corroborate our analytical results with empirical studies on equity markets that are known to be driven by factors.

Conversational AI systems have emerged as key enablers of human-like interactions across diverse sectors. Nevertheless, the balance between linguistic nuance and factual accuracy has proven elusive. In this paper, we first introduce LLMXplorer, a comprehensive tool that provides an in-depth review of over 150 Large Language Models (LLMs), elucidating their myriad implications ranging from social and ethical to regulatory, as well as their applicability across industries. Building on this foundation, we propose a novel functional architecture that seamlessly integrates the structured dynamics of Knowledge Graphs with the linguistic capabilities of LLMs. Validated using real-world AI news data, our architecture adeptly blends linguistic sophistication with factual rigour and further strengthens data security through Role-Based Access Control. This research provides insights into the evolving landscape of conversational AI, emphasizing the imperative for systems that are efficient, transparent, and trustworthy.