Source author record

David Patterson

David Patterson appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

physics.atom-ph physics.chem-ph Machine Learning Genomics Performance Artificial Intelligence Data Structures and Algorithms Distributed, Parallel, and Cluster Computing General Literature hep-ex quant-ph Quantitative Methods

Catalog footprint

What is connected

13works

12topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrink

Machine Learning (ML) workloads have rapidly grown in importance, but raised concerns about their carbon footprint. Four best practices can reduce ML training energy by up to 100x and CO2 emissions up to 1000x. By following best practices, overall ML energy use (across research, development, and production) held steady at <15% of Google's total energy use for the past three years. If the whole ML field were to adopt best practices, total carbon emissions from training would reduce. Hence, we recommend that ML papers include emissions explicitly to foster competition on more than just model quality. Estimates of emissions in papers that omitted them have been off 100x-100,000x, so publishing emissions has the added benefit of ensuring accurate accounting. Given the importance of climate change, we must get the numbers right to make certain that we work on its biggest challenges.

preprint2022arXiv

The Sky Above The Clouds

Technology ecosystems often undergo significant transformations as they mature. For example, telephony, the Internet, and PCs all started with a single provider, but in the United States each is now served by a competitive market that uses comprehensive and universal technology standards to provide compatibility. This white paper presents our view on how the cloud ecosystem, barely over fifteen years old, could evolve as it matures.

preprint2021arXiv

Benchmarking TinyML Systems: Challenges and Direction

Recent advancements in ultra-low-power machine learning (TinyML) hardware promises to unlock an entirely new class of smart applications. However, continued progress is limited by the lack of a widely accepted benchmark for these systems. Benchmarking allows us to measure and thereby systematically compare, evaluate, and improve the performance of systems and is therefore fundamental to a field reaching maturity. In this position paper, we present the current landscape of TinyML and discuss the challenges and direction towards developing a fair and useful hardware benchmark for TinyML workloads. Furthermore, we present our four benchmarks and discuss our selection methodology. Our viewpoints reflect the collective thoughts of the TinyMLPerf working group that is comprised of over 30 organizations.

preprint2020arXiv

MLPerf Training Benchmark

Machine learning (ML) needs industry-standard performance benchmarks to support design and competitive evaluation of the many emerging software and hardware solutions for ML. But ML training presents three unique benchmarking challenges absent from other domains: optimizations that improve training throughput can increase the time to solution, training is stochastic and time to solution exhibits high variance, and software and hardware systems are so diverse that fair benchmarking with the same binary, code, and even hyperparameters is difficult. We therefore present MLPerf, an ML benchmark that overcomes these challenges. Our analysis quantitatively evaluates MLPerf's efficacy at driving performance and scalability improvements across two rounds of results from multiple vendors.

preprint2020arXiv

Ultralong spin coherence times for rubidium atoms in solid parahydrogen via dynamical decoupling

Coherence time is an essential parameter for quantum sensing, quantum information, and quantum computation. In this work, we demonstrate electron spin coherence times as long as 0.1 s for an ensemble of rubidium atoms trapped in a solid parahydrogen matrix. We explore the underlying physics limiting the coherence time. The properties of these matrix isolated atoms are very promising for future applications, including quantum sensing of nuclear spins. If combined with efficient single-atom readout, this would enable NMR and magnetic resonance imaging of single molecules cotrapped with alkali-metal atom quantum sensors within a parahydrogen matrix.

preprint2016arXiv

Continuous probe of cold complex molecules with infrared frequency comb spectroscopy

Cavity-enhanced frequency comb spectroscopy for molecule detection in the mid-infrared powerfully combines high resolution, high sensitivity, and broad spectral coverage. However, this technique, and essentially all spectroscopic methods, is limited in application to relatively small, simple molecules. Here we integrate comb spectroscopy with continuous, cold samples of molecules produced via buffer gas cooling, thus enabling the study of significantly more complex molecules. We report simultaneous gains in resolution, sensitivity, and bandwidth and demonstrate this combined capability with the first rotationally resolved direct absorption spectra in the CH stretch region of several complex molecules. These include nitromethane (CH$_3$NO$_2$), a model system that presents challenging questions to the understanding of large amplitude vibrational motion, as well as several large organic molecules with fundamental spectroscopic and astrochemical relevance, including naphthalene (C$_{10}$H$_8$), adamantane (C$_{10}$H$_{16}$), and hexamethylenetetramine (C$_{6}$N$_4$H$_{12}$). This general spectroscopic tool has the potential to significantly impact the field of molecular spectroscopy, simultaneously improving efficiency, spectral resolution, and specificity by orders of magnitude. This realization could open up new molecular species and new kinetics for precise investigations, including the study of complex molecules, weakly bound clusters, and cold chemistry.

preprint2016arXiv

Direct Time-domain Observation of Conformational Relaxation in Gas-phase Cold Collisions

Cooling molecules in the gas phase is important for precision spectroscopy, cold molecule physics, and physical chemistry. Measurements of conformational relaxation cross sections shed important light on potential energy surfaces and energy flow within a molecule. However, gas-phase conformational cooling has not been previously observed directly. In this work, we directly observe conformational dynamics of 1,2-propanediol in cold (6K) collisions with atomic helium using microwave spectroscopy and buffer-gas cooling. Precise knowledge and control of the collisional environment in the buffer-gas allows us to measure the absolute collision cross-section for conformational relaxation. Several conformers of 1,2-propanediol are investigated and found to have relaxation cross-sections with He ranging from $σ=4.7(3.0)\times10^{-18}\:\mathrm{cm}^{2}$ to $σ>5\times10^{-16}\:\mathrm{cm}^{2}$. Our method is applicable to a broad class of molecules and could be used to provide information about the potential energy surfaces of previously uninvestigated molecules.

preprint2014arXiv

Slow, Continuous Beams of Large Gas Phase Molecules

Cold, continuous, high flux beams of benzonitrile, fluorobenzine, and anisole have been created. Buffer-gas cooling with a cryogenic gas provides the cooling and slow forward beam velocities. The beam of benzonitrile was measured to have a forward velocity peaked at 67 $\pm 5$ m s$^{-1}$, and a continuous flux of $10^{15}$ molecules s$^{-1}$. These beams provide a continuous source for high resolution spectroscopy, and provide an attractive starting point for further spatial manipulation of such molecules, including eventual trapping.

preprint2014arXiv

SMaSH: A Benchmarking Toolkit for Human Genome Variant Calling

Motivation: Computational methods are essential to extract actionable information from raw sequencing data, and to thus fulfill the promise of next-generation sequencing technology. Unfortunately, computational tools developed to call variants from human sequencing data disagree on many of their predictions, and current methods to evaluate accuracy and computational performance are ad-hoc and incomplete. Agreement on benchmarking variant calling methods would stimulate development of genomic processing tools and facilitate communication among researchers. Results: We propose SMaSH, a benchmarking methodology for evaluating human genome variant calling algorithms. We generate synthetic datasets, organize and interpret a wide range of existing benchmarking data for real genomes, and propose a set of accuracy and computational performance metrics for evaluating variant calling methods on this benchmarking data. Moreover, we illustrate the utility of SMaSH to evaluate the performance of some leading single nucleotide polymorphism (SNP), indel, and structural variant calling algorithms. Availability: We provide free and open access online to the SMaSH toolkit, along with detailed documentation, at smash.cs.berkeley.edu.

preprint2013arXiv

Sensitive Chiral Analysis via Microwave Three-wave Mixing

We demonstrate chirality-induced three-wave mixing in the microwave regime, using rotational transitions in cold gas-phase samples of 1,2-propanediol and 1,3-butanediol. We show that bulk three-wave mixing, which can only be realized in a chiral environment, provides a sensitive, species selective probe of enantiomeric excess and is applicable to a broad class of molecules. The doubly resonant condition provides simultaneous identification of species and of handedness, which should allow sensitive chiral analysis even within a complex mixture.

preprint2011arXiv

Faster and More Accurate Sequence Alignment with SNAP

We present the Scalable Nucleotide Alignment Program (SNAP), a new short and long read aligner that is both more accurate (i.e., aligns more reads with fewer errors) and 10-100x faster than state-of-the-art tools such as BWA. Unlike recent aligners based on the Burrows-Wheeler transform, SNAP uses a simple hash index of short seed sequences from the genome, similar to BLAST's. However, SNAP greatly reduces the number and cost of local alignment checks performed through several measures: it uses longer seeds to reduce the false positive locations considered, leverages larger memory capacities to speed index lookup, and excludes most candidate locations without fully computing their edit distance to the read. The result is an algorithm that scales well for reads from one hundred to thousands of bases long and provides a rich error model that can match classes of mutations (e.g., longer indels) that today's fast aligners ignore. We calculate that SNAP can align a dataset with 30x coverage of a human genome in less than an hour for a cost of $2 on Amazon EC2, with higher accuracy than BWA. Finally, we describe ongoing work to further improve SNAP.

preprint2010arXiv

Search for the electric dipole moment of the electron with thorium monoxide

The electric dipole moment of the electron (eEDM) is a signature of CP-violating physics beyond the Standard Model. We describe an ongoing experiment to measure or set improved limits to the eEDM, using a cold beam of thorium monoxide (ThO) molecules. The metastable $H {}^3Δ_1$ state in ThO has important advantages for such an experiment. We argue that the statistical uncertainty of an eEDM measurement could be improved by as much as 3 orders of magnitude compared to the current experimental limit, in a first-generation apparatus using a cold ThO beam. We describe our measurements of the $H$ state lifetime and the production of ThO molecules in a beam, which provide crucial data for the eEDM sensitivity estimate. ThO also has ideal properties for the rejection of a number of known systematic errors; these properties and their implications are described.

preprint2008arXiv

Intense Atomic and Molecular Beams via Neon Buffer Gas Cooling

We realize a continuous guided beam of cold deuterated ammonia with a flux of 3e11 ND3 molecules/s and a continuous free-space beam of cold potassium with a flux of 1e16 K atoms/s. A novel feature of the buffer gas source used to produce these beams is cold neon, which, due to intermediate Knudsen number beam dynamics, produces a forward velocity and low-energy tail that is comparable to much colder helium-based sources. We expect this source to be trivially generalizable to a very wide range of atomic and molecular species with significant vapor pressure below 1000 K. This source has properties that make it a good starting point for laser cooling of molecules or atoms, cold collision studies, trapping, or nonlinear optics in buffer-gas-cooled atomic or molecular gases.