-
Bridging the Skills Gap: Evaluating an AI-Assisted Provider Platform to Support Care Providers with Empathetic Delivery of Protocolized Therapy
Authors:
William R. Kearns,
Jessica Bertram,
Myra Divina,
Lauren Kemp,
Yinzhou Wang,
Alex Marin,
Trevor Cohen,
Weichao Yuwen
Abstract:
Despite the high prevalence and burden of mental health conditions, there is a global shortage of mental health providers. Artificial Intelligence (AI) methods have been proposed as a way to address this shortage, by supporting providers with less extensive training as they deliver care. To this end, we developed the AI-Assisted Provider Platform (A2P2), a text-based virtual therapy interface that…
▽ More
Despite the high prevalence and burden of mental health conditions, there is a global shortage of mental health providers. Artificial Intelligence (AI) methods have been proposed as a way to address this shortage, by supporting providers with less extensive training as they deliver care. To this end, we developed the AI-Assisted Provider Platform (A2P2), a text-based virtual therapy interface that includes a response suggestion feature, which supports providers in delivering protocolized therapies empathetically. We studied providers with and without expertise in mental health treatment delivering a therapy session using the platform with (intervention) and without (control) AI-assistance features. Upon evaluation, the AI-assisted system significantly decreased response times by 29.34% (p=0.002), tripled empathic response accuracy (p=0.0001), and increased goal recommendation accuracy by 66.67% (p=0.001) across both user groups compared to the control. Both groups rated the system as having excellent usability.
△ Less
Submitted 7 January, 2024;
originally announced January 2024.
-
Estimating the Long-term Behavior of Biologically Inspired Agent-based Models
Authors:
Daniel A. Cruz,
Jack Toppen,
Eunbi Park,
Melissa L. Kemp,
Elena S. Dimitrova
Abstract:
An agent-based model (ABM) is a computational model in which the local interactions of autonomous agents with each other and with their environment give rise to global properties within a given domain. As the detail and complexity of these models has grown, so too has the computational expense of running several simulations to perform sensitivity analysis and evaluate long-term model behavior. Her…
▽ More
An agent-based model (ABM) is a computational model in which the local interactions of autonomous agents with each other and with their environment give rise to global properties within a given domain. As the detail and complexity of these models has grown, so too has the computational expense of running several simulations to perform sensitivity analysis and evaluate long-term model behavior. Here, we generalize a framework for mathematically formalizing ABMs to explicitly incorporate features commonly found in biological systems: appearance of agents (birth), removal of agents (death), and locally dependent state changes. We then use our broader framework to extend an approach for estimating long-term behavior without simulations, specifically changes in population densities over time. The approach is probabilistic and relies on treating the discrete, incremental update of an ABM via "time steps" as a Markov process to generate expected values for agents at each time step. As case studies, we apply our extensions to both a simple ABM based on the Game of Life and a published ABM of rib development in vertebrates.
△ Less
Submitted 30 November, 2022; v1 submitted 1 November, 2022;
originally announced November 2022.
-
Democratising Risk: In Search of a Methodology to Study Existential Risk
Authors:
Carla Zoe Cremer,
Luke Kemp
Abstract:
Studying potential global catastrophes is vital. The high stakes of existential risk studies (ERS) necessitate serious scrutiny and self-reflection. We argue that existing approaches to studying existential risk are not yet fit for purpose, and perhaps even run the risk of increasing harm. We highlight general challenges in ERS: accommodating value pluralism, crafting precise definitions, developi…
▽ More
Studying potential global catastrophes is vital. The high stakes of existential risk studies (ERS) necessitate serious scrutiny and self-reflection. We argue that existing approaches to studying existential risk are not yet fit for purpose, and perhaps even run the risk of increasing harm. We highlight general challenges in ERS: accommodating value pluralism, crafting precise definitions, developing comprehensive tools for risk assessment, dealing with uncertainty, and accounting for the dangers associated with taking exceptional actions to mitigate or prevent catastrophes. The most influential framework for ERS, the 'techno-utopian approach' (TUA), struggles with these issues and has a unique set of additional problems: it unnecessarily combines the study of longtermism and longtermist ethics with the study of extinction, relies on a non-representative moral worldview, uses ambiguous and inadequate definitions, fails to incorporate insights from risk assessment in relevant fields, chooses arbitrary categorisations of risk, and advocates for dangerous mitigation strategies. Its moral and empirical assumptions might be particularly vulnerable to securitisation and misuse. We suggest several key improvements: separating the study of extinction ethics (ethical implications of extinction) and existential ethics (the ethical implications of different societal forms), from the analysis of human extinction and global catastrophe; drawing on the latest developments in risk assessment literature; diversifying the field, and; democratising its policy recommendations.
△ Less
Submitted 27 December, 2021;
originally announced January 2022.
-
Should Artificial Intelligence Governance be Centralised? Design Lessons from History
Authors:
Peter Cihon,
Matthijs M. Maas,
Luke Kemp
Abstract:
Can effective international governance for artificial intelligence remain fragmented, or is there a need for a centralised international organisation for AI? We draw on the history of other international regimes to identify advantages and disadvantages in centralising AI governance. Some considerations, such as efficiency and political power, speak in favour of centralisation. Conversely, the risk…
▽ More
Can effective international governance for artificial intelligence remain fragmented, or is there a need for a centralised international organisation for AI? We draw on the history of other international regimes to identify advantages and disadvantages in centralising AI governance. Some considerations, such as efficiency and political power, speak in favour of centralisation. Conversely, the risk of creating a slow and brittle institution speaks against it, as does the difficulty in securing participation while creating stringent rules. Other considerations depend on the specific design of a centralised institution. A well-designed body may be able to deter forum shopping and ensure policy coordination. However, forum shopping can be beneficial and a fragmented landscape of institutions can be self-organising. Centralisation entails trade-offs and the details matter. We conclude with two core recommendations. First, the outcome will depend on the exact design of a central institution. A well-designed centralised regime covering a set of coherent issues could be beneficial. But locking-in an inadequate structure may pose a fate worse than fragmentation. Second, for now fragmentation will likely persist. This should be closely monitored to see if it is self-organising or simply inadequate.
△ Less
Submitted 10 January, 2020;
originally announced January 2020.
-
Interspecies thermalization in an ultracold mixture of Cs and Yb in an optical trap
Authors:
A. Guttridge,
S. A. Hopkins,
S. L. Kemp,
Matthew D. Frye,
Jeremy M. Hutson,
Simon L. Cornish
Abstract:
We present measurements of interspecies thermalization between ultracold samples of $^{133}$Cs and either $^{174}$Yb or $^{170}$Yb. The two species are trapped in a far-off-resonance optical dipole trap and $^{133}$Cs is sympathetically cooled by Yb. We extract effective interspecies thermalization cross sections by fitting the thermalization measurements to a rate equation model, giving…
▽ More
We present measurements of interspecies thermalization between ultracold samples of $^{133}$Cs and either $^{174}$Yb or $^{170}$Yb. The two species are trapped in a far-off-resonance optical dipole trap and $^{133}$Cs is sympathetically cooled by Yb. We extract effective interspecies thermalization cross sections by fitting the thermalization measurements to a rate equation model, giving $σ_{\mathrm{Cs^{174}Yb}} = \left(5 \pm 2\right) \times 10^{-13} \, \mathrm{cm^{2}}$ and $σ_{\mathrm{Cs^{170}Yb}} = \left(18 \pm 8\right) \times 10^{-13} \, \mathrm{cm^{2}}$. We perform quantum scattering calculations of the thermalization cross sections and optimize the CsYb interaction potential to reproduce the measurements. We predict scattering lengths for all isotopic combinations of Cs and Yb. We also demonstrate the independent production of $^{174}$Yb and $^{133}$Cs Bose-Einstein condensates using the same optical dipole trap, an important step towards the realization of a quantum-degenerate mixture of the two species.
△ Less
Submitted 16 July, 2017; v1 submitted 11 April, 2017;
originally announced April 2017.
-
Direct loading of a large Yb MOT on the $^{1}S_{0} \rightarrow \, ^{3}P_{1}$ transition
Authors:
A Guttridge,
S A Hopkins,
S L Kemp,
D Boddy,
R Freytag,
M P A Jones,
M R Tarbutt,
E A Hinds,
S L Cornish
Abstract:
We report a robust technique for laser frequency stabilisation that enables the reproducible loading of in excess of 10$^{9}$ Yb atoms from a Zeeman slower directly into a magneto-optical trap (MOT) operating on the $^{1}S_{0} \rightarrow \, ^{3}P_{1}$ transition, without the need for a first stage MOT on the $^{1}S_{0} \rightarrow \, ^{1}P_{1}$ transition. We use a simple atomic beam apparatus to…
▽ More
We report a robust technique for laser frequency stabilisation that enables the reproducible loading of in excess of 10$^{9}$ Yb atoms from a Zeeman slower directly into a magneto-optical trap (MOT) operating on the $^{1}S_{0} \rightarrow \, ^{3}P_{1}$ transition, without the need for a first stage MOT on the $^{1}S_{0} \rightarrow \, ^{1}P_{1}$ transition. We use a simple atomic beam apparatus to generate narrow fluorescence signals on both the 399 nm $^{1}S_{0} \rightarrow \, ^{1}P_{1}$ transition used for the Zeeman slower and the 556 nm $^{1}S_{0} \rightarrow \, ^{3}P_{1}$ transition. We present in detail the methods for obtaining spectra with a high signal-to-noise ratio and demonstrate error signals suitable for robust frequency stabilisation. Finally we demonstrate the stability and precision of our technique through sensitive measurements of the gravitational sag of the Yb MOT as a function of the intensity of the laser cooling beams, which are in good agreement with theory. These results will be important for efficient loading of the atoms into an optical dipole trap.
△ Less
Submitted 7 June, 2016; v1 submitted 21 December, 2015;
originally announced December 2015.
-
Production and characterization of a dual species magneto-optical trap of cesium and ytterbium
Authors:
S. L. Kemp,
K. L. Butler,
R. Freytag,
S. A. Hopkins,
E. A. Hinds,
M. R. Tarbutt,
S. L. Cornish
Abstract:
We describe an apparatus designed to trap and cool a Yb and Cs mixture. The apparatus consists of a dual species effusive oven source, dual species Zeeman slower, magneto-optical traps in a single ultra-high vacuum science chamber, and the associated laser systems. The dual species Zeeman slower is used to load sequentially the two species into their respective traps. Its design is flexible and ma…
▽ More
We describe an apparatus designed to trap and cool a Yb and Cs mixture. The apparatus consists of a dual species effusive oven source, dual species Zeeman slower, magneto-optical traps in a single ultra-high vacuum science chamber, and the associated laser systems. The dual species Zeeman slower is used to load sequentially the two species into their respective traps. Its design is flexible and may be adapted for other experiments with different mixtures of atomic species. The apparatus provides excellent optical access and can apply large magnetic bias fields to the trapped atoms. The apparatus regularly produces 10${}^{8}$ Cs atoms at 13.3 $μ$K in an optical molasses, and 10${}^{9}$ Yb atoms cooled to 22 $μ$K in a narrowband magneto-optical trap.
△ Less
Submitted 3 July, 2015;
originally announced July 2015.
-
An analytical model of Faraday rotation in hot alkali metal vapours
Authors:
Stefan L Kemp,
Ifan G Hughes,
Simon L Cornish
Abstract:
We report a thorough investigation into the absorptive and dispersive properties of hot caesium vapour, culminating in the development of a simple analytical model for off-resonant Faraday rotation. The model, applicable to all hot alkali metal vapours, is seen to predict the rotation observed in caesium, at temperatures as high as 115 $^{\circ}$C, to within 1% accuracy for probe light detuned by…
▽ More
We report a thorough investigation into the absorptive and dispersive properties of hot caesium vapour, culminating in the development of a simple analytical model for off-resonant Faraday rotation. The model, applicable to all hot alkali metal vapours, is seen to predict the rotation observed in caesium, at temperatures as high as 115 $^{\circ}$C, to within 1% accuracy for probe light detuned by greater than 2 GHz from the $D_{2}$ lines. We also demonstrate the existence of a weak probe intensity limit, below which the effect of hyperfine pumping is negligible. Following the identification of this regime we validate a more comprehensive model for the absorption and dispersion in the vicinity of the $D_{2}$ lines, implemented in the form of a computer code. We demonstrate the ability of this model to predict Doppler-broadened spectra to within 0.5% rms deviation for temperatures up to 50 $^{\circ}$C.
△ Less
Submitted 7 July, 2011; v1 submitted 6 July, 2011;
originally announced July 2011.
