19 research outputs found
Pushdown Automata and Context-Free Grammars in Bisimulation Semantics
No full textThe Turing machine models an old-fashioned computer, that does not interact with the user or with other computers, and only does batch processing. Therefore, we came up with a Reactive Turing Machine that does not have these shortcomings. In the Reactive Turing Machine, transitions have labels to give a notion of interactivity. In the resulting process graph, we use bisimilarity instead of language equivalence.
Subsequently, we considered other classical theorems and notions from automata theory and formal languages theory. In this paper, we consider the classical theorem of the correspondence between pushdown automata and context-free grammars. By changing the process operator of sequential composition to a sequencing operator with intermediate acceptance, we get a better correspondence in our setting. We find that the missing ingredient to recover the full correspondence is the addition of a notion of state awareness
Counter-Intuitive Effects of <italic>Q</italic>-Learning Exploration in a Congestion Dilemma
No full textExploration is an integral part of learning dynamics which allows algorithms to search a space of solutions. When many algorithms simultaneously explore, this can lead to counter-intuitive effects. This paper contributes an analysis of the influence that exploration has on a multi-agent system of -learners in a famous congestion dilemma, the Braess paradox. I find ranges of the exploration rate for which -greedy -learners show chaotic and oscillatory dynamics which do not converge, and yield better than Nash equilibrium results. I decouple the dynamics endogenous to -learning from the exogenous exploration rate , and find that -learners implicitly coordinate with low exploration rates , but are disrupted in their coordination for larger exploration rates \epsilon > 0.1 . The best implicit coordination leads to a 20% reduction in average travel times which approaches the social optimum. I discuss how our results may inform multi-agent algorithm design, fit within a cognitive science perspective of cognitive noise during learning, and provide a mechanistic hypothesis for the lack of empirical evidence of the Braess Paradox in traffic systems
Sequential combination chemotherapy of high-grade non-Hodgkin's lymphoma with 5-fluorouracil, methotrexate, cytosine-arabinoside, cyclophosphamide, doxorubicin, vincristine, and prednisone (F-MACHOP)
No full textAn intensive treatment program was developed to achieve durable remissions in a high proportion of previously untreated patients with advanced stages of diffuse high-grade non-Hodgkin's lymphoma (NHL). Fifty-six patients (15-68 years) received a course of F-MACHOP (5-fluorouracil, methotrexate, cytosine-arabinoside, cyclophosphamide, doxorubicin, vincristine, and prednisone) every 3-4 weeks for 6 courses. Cycle active drugs were sequentially administered to expose rapidly proliferating tumor cells to the synergistic effects of these agents throughout the cell cycle. Forty-three patients achieved complete remission (77%) and 80% of the complete responders are projected to be alive and disease-free at 4 1/2 years (median follow-up 33 months). Up to 70% of all patients are predicted to be alive at 5 years. Bulky tumor, 'B'-symptoms and lymphoblastic histology were poor prognostic factors, particularly when associated with clinically detectable disease after three courses. Toxicity included transitory myelodepression in most patients (2 septic deaths). This protocol provides effective and tolerable therapy for the majority of patients with advanced stages of diffuse aggressive NHL
Gift-giving, consumption and the female court in sixteenth-century Italy
No full textPhDThe subject of my research is the female consort and her court. I focus on three Austrian
Archduchesses: Giovanna, Barbara and Eleonora Habsburg who came down to Italy in
the second half of the sixteenth century and married into the ducal houses of Florence,
Ferrara and Mantua respectively. My thesis compares the structures, roles and
relationships in these three contemporary female courts, and analyses the consorts’
reliance on personal consumption, gift-giving and patronage activities to assert their
power, position and identity. My research is primarily based on the unpublished letters
and accounts preserved in the three state archives of Florence, Modena (which contains
the Este archive) and Mantua.
My thesis starts with a background chapter on the history of the three Duchesses,
and then turns to address each Duchess’s financial situation, the organisation of her
court, her attitude to her husband and her new family and the particular circumstances
of her life. This chapter sheds new light on the position of the consort, and sets the stage
for the exploration of her patronage and consumption. My first case-study focuses on
clothing. I examine the Duchesses’ choices in dressing themselves and their courts and
analyse their treatment of clothing as a valuable visual language. My second case-study
focuses on the gifts of food that were sent to and from the Duchesses. I discuss their
function as items of relatively small economic value in the creation of patronage
relationships and in the process of social and political mediation.
The central tenet in my case-studies is that objects could act as coded messages,
with multiple meanings which can be dissected by studying owner, receiver, means of
transmission and the type of object itself. My approach employs material culture as a
means for enriching current knowledge of a particularly under-researched subject: the
female consort
Capital as Artificial Intelligence
No full textWe gather many perspectives on Capital and synthesize their commonalities. We provide a characterization of Capital as a historical agential system and propose a model of Capital using tools from computer science. Our model consists of propositions which, if satisfied by a specific grounding, constitute a valid model of Capital. We clarify the manners in which Capital can evolve. We claim that, when its evolution is driven by quantitative optimization processes, Capital can possess qualities of Artificial Intelligence. We find that Capital may not uniquely represent meaning, in the same way that optimization is not intentionally meaningful. We find that Artificial Intelligences like modern day Large Language Models are a part of Capital. We link our readers to a web-interface where they can interact with a part of Capital
Dynamic value alignment through preference aggregation of multiple objectives
No full textThe development of ethical AI systems is currently geared toward setting
objective functions that align with human objectives. However, finding such
functions remains a research challenge, while in RL, setting rewards by hand is
a fairly standard approach. We present a methodology for dynamic value
alignment, where the values that are to be aligned with are dynamically
changing, using a multiple-objective approach. We apply this approach to extend
Deep -Learning to accommodate multiple objectives and evaluate this method
on a simplified two-leg intersection controlled by a switching agent.Our
approach dynamically accommodates the preferences of drivers on the system and
achieves better overall performance across three metrics (speeds, stops, and
waits) while integrating objectives that have competing or conflicting actions
Pushdown automata and context-free grammars in bisimulation semantics
No full textThe Turing machine models an old-fashioned computer, that does not interact with the user or with other computers, and only does batch processing. Therefore, we came up with a Reactive Turing Machine that does not have these shortcomings. In the Reactive Turing Machine, transitions have labels to give a notion of interactivity. In the resulting process graph, we use bisimilarity instead of language equivalence. Subsequently, we considered other classical theorems and notions from automata theory and formal languages theory. In this paper, we consider the classical theorem of the correspondence between pushdown automata and context-free grammars. By changing the process operator of sequential composition to a sequencing operator with intermediate acceptance, we get a better correspondence in our setting. We find that the missing ingredient to recover the full correspondence is the addition of a notion of state awareness
Pushdown automata and context-free grammars in bisimulation semantics
No full textThe Turing machine models an old-fashioned computer, that does not interact with the user or with other computers, and only does batch processing. Therefore, we came up with a Reactive Turing Machine that does not have these shortcomings. In the Reactive Turing Machine, transitions have labels to give a notion of interactivity. In the resulting process graph, we use bisimilarity instead of language equivalence. Subsequently, we considered other classical theorems and notions from automata theory and formal languages theory. In this paper, we consider the classical theorem of the correspondence between pushdown automata and context-free grammars. By changing the process operator of sequential composition to a sequencing operator with intermediate acceptance, we get a better correspondence in our setting. We find that the missing ingredient to recover the full correspondence is the addition of a notion of state awareness
The man behind the curtain: appropriating fairness in AI
No full textOur goal in this paper is to establish a set of criteria for understanding the meaning and sources of attributing (un)fairness to AI algorithms. To do so, we first establish that (un)fairness, like other normative notions, can be understood in a proper primary sense and in secondary senses derived by analogy. We argue that AI algorithms cannot be said to be (un)fair in the proper sense due to a set of criteria related to normativity and agency. However, we demonstrate how and why AI algorithms can be qualified as (un)fair by analogy and explore the sources of this (un)fairness and the associated problems of responsibility assignment. We conclude that more user-driven AI approaches could alleviate some of these difficulties
